CN115429326A

CN115429326A - Ultrasonic imaging method and ultrasonic imaging equipment

Info

Publication number: CN115429326A
Application number: CN202210963825.8A
Authority: CN
Inventors: 贾洪飞; 叶焜; 梁天柱; 林穆清; 邹耀贤
Original assignee: Shenzhen Mindray Bio Medical Electronics Co Ltd
Current assignee: Shenzhen Mindray Bio Medical Electronics Co Ltd
Priority date: 2018-12-28
Filing date: 2018-12-28
Publication date: 2022-12-06
Also published as: CN111374712B; CN115429325A; CN111374712A

Abstract

The embodiment of the application discloses an ultrasonic imaging method, which is applied to ultrasonic imaging equipment and comprises the following steps: acquiring three-dimensional volume data of a first tissue to be detected; identifying three-dimensional volume data of a vertebral column of a fetus and three-dimensional volume data of ribs of the fetus from the three-dimensional volume data of the first tissue to be detected; rendering the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs to obtain a three-dimensional ultrasonic image of a fetal rib structure; marking the fetal ribs in the three-dimensional ultrasound image of the fetal rib structure; and outputting the marked three-dimensional ultrasonic image of the fetal rib structure. The embodiment of the application discloses ultrasonic imaging equipment.

Description

Ultrasonic imaging method and ultrasonic imaging equipment

The present disclosure is proposed based on the chinese patent application having an application number of 201811623609.9, an application date of 2018, 12 and 28, and an invention name of "an ultrasonic imaging method and an ultrasonic imaging apparatus", and a divisional case is proposed within the scope described in the chinese patent application, the entire contents of which are incorporated herein by reference.

Technical Field

The embodiment of the application relates to medical diagnosis technology, in particular to but not limited to an ultrasonic imaging method and an ultrasonic imaging device.

Background

Prenatal ultrasound examination is one of the most important items that an expectant mother must examine during pregnancy, and the main functions of the prenatal ultrasound examination include determining the age of a fetus, analyzing the development condition of the fetus, detecting the abnormality or abnormality of the fetus, taking pictures and dynamic video of the fetus and the like. The abnormal number of the ribs of the fetus and the position of the conical end of the spinal cord are one of the items of prenatal ultrasound screening and detecting.

Abnormal fetal rib (or vertebra) number (or lesion) is usually associated with fetal chromosomal abnormalities and clinical syndromes, such as trisomy syndrome, dysthymia thoraco syndrome, dysplasia type I, and the like. The location of the conical end of the spinal cord during fetal periods is of great importance for early screening of spinal cord tethering syndromes. The nerve damage syndrome caused by the fact that the conical end position of the spinal cord of a fetus does not change constantly with the change of gestational age, the spinal cord is longitudinally pulled due to congenital reasons, and the conical end position of the spinal cord descends when the spinal cord is changed pathologically is called spinal tethering syndrome. Failure to detect these congenital anomalies in the prenatal examination would impose a tremendous mental and economic burden on the patient's family and society, and even cause the patient to die during neonatal period due to dyspnea caused by thoracic deformity.

Ultrasound is a safe, convenient, noninvasive and high-repeatability imaging technology, and two-dimensional and three-dimensional ultrasound can characteristically reflect the abnormalities of fetal ribs, fetal spines and vertebral bodies, so that the ultrasound becomes a preferred mode for doctors to diagnose the abnormalities of the fetal ribs. The two-dimensional ultrasound can inspect the ribs and the spinal cords of the fetus from a sagittal plane, a transverse plane and a coronal plane, so that the conditions of the ribs, the spinal cords and the spinal cords of the fetus can be comprehensively known; the three-dimensional ultrasound can overcome the problems of lack of two-dimensional ultrasound space sense, poor fidelity, difficult positioning and the like in fetal rib and spinal cord examination. However, at present, doctors still need a lot of purely manual operations when performing fetal rib and spinal cord examinations by using three-dimensional ultrasound, and the examinations still have the following pain points:

the fetus can be in various positions during examination, and the median sagittal plane of the vertebral column of the fetus needs to be taken as a starting plane when the three-dimensional ultrasonic volume is acquired. After obtaining the three-dimensional data of the ribs and the spinal column of the fetus, a doctor needs to have very deep understanding on the three-dimensional space, and can select a region of interest (VOI) to cut under three-dimensional ultrasound through multiple times of manual rotation and translation geometric operations and Virtual display (VR) to determine the accurate position of the ribs with deficiency and lesion of the fetus or determine the accurate position of the spinal cone.

Disclosure of Invention

In view of this, the present application provides an ultrasound imaging method and an ultrasound imaging apparatus.

The technical scheme of the embodiment of the application is realized as follows:

in one aspect, an embodiment of the present application provides an ultrasound imaging method, which is applied to an ultrasound imaging apparatus, and the method includes:

acquiring three-dimensional volume data of a first tissue to be detected; identifying three-dimensional volume data of a vertebral column of a fetus and three-dimensional volume data of ribs of the fetus from the three-dimensional volume data of the first tissue to be detected; rendering the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs to obtain a three-dimensional ultrasonic image of a fetal rib structure; marking the fetal ribs in the three-dimensional ultrasound image of the fetal rib structure; and outputting the marked three-dimensional ultrasonic image of the fetal rib structure.

In one aspect, an embodiment of the present application provides an ultrasound imaging method, which is applied to an ultrasound imaging apparatus, and the method includes: acquiring three-dimensional volume data of a second tissue to be detected; identifying three-dimensional volume data of spinal cones and three-dimensional volume data of lumbar vertebrae from the three-dimensional volume data of the second tissue to be detected; rendering the three-dimensional volume data of the spinal cord cone and the three-dimensional volume data of the lumbar vertebra to obtain a three-dimensional ultrasonic image of a vertebra structure; marking the spinal cone in a three-dimensional ultrasound image of the vertebral structure; outputting a three-dimensional ultrasound image of the marked vertebral structure.

In one aspect, an embodiment of the present application provides an ultrasound imaging method, which is applied to an ultrasound imaging apparatus, and the method includes: acquiring three-dimensional volume data of a fetus; identifying three-dimensional volume data of the ribs of the fetus from the three-dimensional volume data of the fetus based on the characteristics of the ribs of the fetus; according to the identified three-dimensional volume data of the fetal ribs, obtaining a first plane or a first curved surface which passes through at least two ribs in the three-dimensional volume data of the fetal ribs and is parallel to or coincided with the arrangement surface of a plurality of fetal ribs in the three-dimensional volume data of the fetal ribs and/or obtaining a second plane or a second curved surface which passes through at least one rib in the three-dimensional volume data of the fetal ribs and is intersected with the arrangement surface of the plurality of fetal ribs in the three-dimensional volume data of the fetal ribs; obtaining an image on the first plane or the first curved surface and/or obtaining an image on the second plane or the second curved surface according to the three-dimensional volume data of the identified ribs of the fetus; and displaying the image on the first plane or the first curved surface as a two-dimensional image and/or displaying the image on the second plane or the second curved surface as a two-dimensional image.

In one aspect, in an embodiment of the present application, there is provided an ultrasound imaging method applied to an ultrasound imaging apparatus, the method including: acquiring three-dimensional volume data of a fetus; identifying a spinal cone region from three-dimensional volume data of a fetus based on characteristics of the spinal cone of the fetus; determining the position of the spinal conical region according to the identified spinal conical region; displaying the location of the spinal conical region.

In one aspect, an embodiment of the present application provides an ultrasound imaging apparatus, including:

a probe;

the transmitting circuit is used for exciting the probe to transmit ultrasonic waves to the first tissue to be detected;

the receiving circuit receives an ultrasonic echo returned from the first tissue to be detected through the probe to obtain an ultrasonic echo signal;

a processor that processes the ultrasound echo signals to obtain a three-dimensional ultrasound image of the marked fetal rib structure;

a display for displaying the marked three-dimensional ultrasonic image of the fetal rib structure;

wherein the processor further performs the steps of:

acquiring three-dimensional volume data of a first tissue to be detected according to the ultrasonic echo information;

identifying three-dimensional volume data of a vertebral column of a fetus and three-dimensional volume data of ribs of the fetus from the three-dimensional volume data of the first tissue to be detected;

rendering the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs to obtain a three-dimensional ultrasonic image of the fetal rib structure;

marking the fetal ribs in the three-dimensional ultrasound image of the fetal rib structure;

and outputting a three-dimensional ultrasonic image of the marked fetal rib structure.

a probe;

the transmitting circuit is used for exciting the probe to transmit ultrasonic waves to a second tissue to be detected;

the receiving circuit receives the ultrasonic echo returned from the second tissue to be detected through the probe so as to obtain an ultrasonic echo signal;

a processor that processes the ultrasound echo signals to obtain a three-dimensional ultrasound image of the marked vertebral structure;

a display that displays a three-dimensional ultrasound image of the marked vertebral structure;

wherein the processor further performs the steps of:

acquiring three-dimensional volume data of a second tissue to be detected according to the ultrasonic echo information;

identifying three-dimensional volume data of spinal cord cones and three-dimensional volume data of lumbar vertebrae from the second three-dimensional volume data;

rendering the three-dimensional data of the spinal cord cone and the three-dimensional data of the lumbar vertebra to obtain a three-dimensional ultrasonic image of the vertebra structure;

marking the spinal cone in a three-dimensional ultrasound image of the vertebral structure;

outputting a three-dimensional ultrasound image of the marked vertebral structure.

In the embodiment of the application, three-dimensional data of a fetal spine and three-dimensional data of a fetal rib are identified from three-dimensional data of a first tissue to be detected, or three-dimensional data of a spinal cone and three-dimensional data of a lumbar vertebra are identified from three-dimensional data of a second tissue to be detected, the identified three-dimensional data are rendered, a three-dimensional ultrasonic image of a fetal rib structure or a three-dimensional ultrasonic image of the spinal column are obtained, and the fetal rib or the spinal cone is marked in the displayed three-dimensional ultrasonic image; therefore, the positions of the fetal ribs or the spinal cord cones are identified and detected from the three-dimensional data, the number of the fetal ribs is automatically counted, the relative positions of the spinal cord cones and the lumbar vertebrae are calculated, the workflow of fetal rib or spinal cord cone position detection is simplified, the detection efficiency is improved, a doctor is liberated from complex time-consuming operation, the dependency of fetal rib or spinal cord cone position detection on the detection doctor technology is reduced, and the detection efficiency is improved.

Drawings

Fig. 1 is a first schematic structural diagram of an ultrasonic imaging apparatus provided in an embodiment of the present application;

fig. 2 is a first schematic flowchart of an ultrasound imaging method provided in an embodiment of the present application;

fig. 3 is a schematic flowchart of a second ultrasound imaging method provided in an embodiment of the present application;

fig. 4 is a schematic flowchart three of an ultrasound imaging method provided in an embodiment of the present application;

fig. 5 is a fourth schematic flowchart of an ultrasonic imaging method provided in an embodiment of the present application;

fig. 6 is a fifth schematic flowchart of an ultrasound imaging method provided in an embodiment of the present application;

fig. 7 is a schematic representation of a fetal rib;

FIG. 8 is a diagram illustrating a manually drawn anatomical trajectory in the related art;

FIG. 9 is a schematic view of the spinal cord cone;

fig. 10 is a schematic structural diagram of a second component of an ultrasound imaging apparatus provided in an embodiment of the present application;

fig. 11 is a schematic diagram of the arrangement sequence of fetal ribs in the fetal rib structure;

fig. 12 is a schematic view of the alignment of the fetal ribs and spine in the fetal rib structure of the present application;

FIG. 13 is a cross-sectional view of all ribs in the present embodiment;

FIG. 14 is a cross-sectional view of a rib in accordance with an embodiment of the present invention;

fig. 15 is a schematic diagram of the three-dimensional skeleton effect of the fetal rib structure in the embodiment of the present application;

fig. 16 is a schematic diagram illustrating the labeling effect of the spinal cord cone in the VR image in the embodiment of the application.

Detailed Description

The present application will be described in further detail below with reference to the accompanying drawings and examples. It should be understood that the examples provided herein are merely illustrative of the present application and are not to be construed as limiting the present application. In addition, the following embodiments are provided as partial embodiments configured to implement the present application, and not to provide all embodiments for implementing the present application, and the technical solutions described in the embodiments of the present application may be implemented in any combination without conflict.

Fig. 1 is a schematic structural block diagram of an ultrasound imaging apparatus 10 in an embodiment of the present application. The ultrasound imaging device 10 may include a probe 100, a transmit circuit 101, a transmit/receive select switch 102, a receive circuit 103, a beam forming circuit 104, a processor 105, and a display 106. The transmit circuit 101 may excite the probe 100 to transmit ultrasound waves to the target object. The receiving circuit 103 may receive an ultrasonic echo returned from the target object through the probe 100, thereby obtaining an ultrasonic echo signal. The ultrasonic echo signal is subjected to beamforming processing by the beamforming circuit 104, and then sent to the processor 105. The processor 105 processes the ultrasound echo signals to obtain an ultrasound image of the target object. The ultrasound images obtained by the processor 105 may be stored in the memory 107. These ultrasound images may be displayed on the display 106.

The target object comprises at least one of a first tissue to be detected and a second tissue to be detected. The first tissue to be tested comprises a fetal rib structure, and the second tissue to be tested comprises a spine.

The embodiment of the present application provides an ultrasound imaging method, which is applied to an ultrasound imaging apparatus shown in fig. 1, and as shown in fig. 2, the method includes:

s201, acquiring three-dimensional volume data of a first tissue to be detected;

the first tissue to be tested comprises: fetal rib structures and tissues outside of fetal rib structures, such as: amniotic fluid region, placenta, uterine wall, etc.

The doctor can scan the pregnant woman through the probe to acquire the three-dimensional data of the first tissue to be detected.

Here, the fetal rib structure belongs to a high echo region, and is displayed in high gray in the ultrasound image.

S202, identifying three-dimensional volume data of a spine of a fetus and three-dimensional volume data of ribs of the fetus from the three-dimensional volume data of the first tissue to be detected;

after obtaining the three-dimensional volume data of the first tissue to be measured based on S201, the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal rib are identified from the obtained three-dimensional volume data. Here, the identification of the three-dimensional volume data of the spinal column of the fetus and the three-dimensional volume data of the ribs of the fetus includes at least one of the following two identification modes:

the identification method comprises the steps of firstly, taking the whole fetal rib structure as an identification object to identify three-dimensional volume data of a fetal rib structure from three-dimensional volume data of a first tissue to be detected, and then identifying the three-dimensional volume data of a fetal rib in the fetal rib structure from the three-dimensional volume data of the fetal rib structure.

And the second identification mode is to directly identify the three-dimensional volume data of the ribs of the fetus from the three-dimensional volume data of the first tissue to be detected by taking the ribs of the fetus as an identification object.

In the first identification mode, identifying the three-dimensional volume data of the rib structure of the fetus from the three-dimensional volume data of the first tissue to be detected; segmenting the three-dimensional volume data of the fetal rib structure from the three-dimensional volume data of the first tissue to be detected; and identifying three-dimensional volume data of a spine of the fetus and three-dimensional volume data of ribs of the fetus from the three-dimensional volume data of the rib structure of the fetus.

In the second identification mode, based on the first rib detection model, different fetal ribs or fetal spines are respectively used as different identification objects, and the three-dimensional volume data of the fetal spines and the three-dimensional volume data of the fetal ribs are identified from the three-dimensional volume data of the first tissue to be detected.

S203, rendering the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs to obtain a three-dimensional ultrasonic image of a fetal rib structure;

after the three-dimensional volume data of the fetal ribs and the three-dimensional volume data of the spine included in the first tissue to be detected are identified in S202, the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs are subjected to three-dimensional rendering, so as to obtain a three-dimensional ultrasonic image of the fetal rib structure.

When the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs are rendered in a three-dimensional mode, all data except the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs in the three-dimensional volume data of the first tissue to be detected are emptied, and the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs are rendered in a three-dimensional light perspective mode to form a three-dimensional ultrasonic image of a fetal rib structure.

It should be noted that the rendering manner may be various, and the rendering manner is not limited in any way in the embodiment of the present application.

S204, marking the fetal ribs in the three-dimensional ultrasonic image of the fetal rib structure;

and determining the position of the three-dimensional data of the fetal rib in the three-dimensional ultrasonic image of the fetal rib structure in the step S203 according to the position of the three-dimensional data of the fetal rib identified in the step S202 in the three-dimensional volume data of the first tissue to be detected, and marking the fetal rib in the fetal rib structure according to the determined position of the fetal rib to obtain the marked three-dimensional ultrasonic image of the fetal rib structure.

In marking the fetal ribs, the rib identification of one or more fetal ribs may be marked: such as: T1-T12, wherein T1 to T12 characterize the 1 st to 12 th fetal ribs, respectively.

In practical application, when the fetal rib is marked, whether the abnormal fetal rib with abnormal shape or pathological changes exists in the fetal rib structure can be judged according to the rib characteristics such as the shape characteristic and the gray characteristic of the fetal rib. If the abnormal fetal ribs are included in the fetal rib structure, the abnormal fetal ribs can be identified through the abnormal identification, and abnormal information of the abnormal fetal ribs relative to the normal fetal ribs is determined.

And S205, outputting the marked three-dimensional ultrasonic image of the fetal rib structure.

At this time, the contents displayed on the display of the ultrasound imaging apparatus are: the three-dimensional ultrasound image of the fetal rib structure marked with the fetal ribs enables a user to visually view the fetal rib structure of the fetus and distinguish the fetal ribs in the fetal rib structure.

In an embodiment, the method further comprises: straightening the three-dimensional data of the fetal rib to obtain straightened rib three-dimensional data; straightening the three-dimensional volume data of the fetal spine to obtain straightened spine three-dimensional volume data; performing plane fitting on a straightened rib in the straightened rib three-dimensional volume data and a straightened fetal spine in the straightened spine three-dimensional volume data to obtain a first plane; and obtaining image data on the first plane according to the straightened rib three-dimensional volume data and the straightened spine three-dimensional volume data to obtain a coronal plane image of the fetal rib structure.

After the three-dimensional volume data of the fetal ribs are determined, straightening the fetal ribs according to the three-dimensional volume data of the fetal ribs, and straightening the bent fetal ribs to obtain the straightened ribs. After the three-dimensional volume data of the fetal spine are determined, straightening the fetal ribs according to the three-dimensional volume data of the fetal spine, straightening the bent fetal spine, and obtaining the straightened fetal spine. At the moment, a first plane is determined by solving a mathematical equation or a plane fitting method similar to a least square method or Hough transformation and the like, and the first plane is a coronal plane in which the longitudinal axis of the straightened rib and the straightened fetal spine are located. And acquiring three-dimensional volume data of each fetal rib and three-dimensional volume data of a fetal spine on the first plane to obtain a coronal plane of the fetal rib structure.

In practical application, after the first plane is determined, the image data of each pixel point on the first plane is obtained according to the three-dimensional volume data of each straightened rib and the three-dimensional volume data of the straightened fetus spine, so that the coronal plane image of the straightened fetus rib structure is obtained.

In an embodiment, the method further comprises: determining a target fetal rib from the ribs of the fetal rib structure; straightening the three-dimensional volume data of the target fetal rib to obtain straightened target fetal rib three-dimensional volume data; straightening the three-dimensional volume data of the fetal spine to obtain straightened spine three-dimensional volume data; taking a plane which is positioned on the same plane with the straightening target ribs and is vertical to the straightening fetus spine as a second plane; and obtaining image data on a second plane according to the three-dimensional volume data of the straightened target fetal rib and the straightened spine three-dimensional volume data to obtain the cross section of the target fetal rib.

Here, the target fetal rib may include all of the fetal ribs in the fetal rib structure, or may be a part of the fetal ribs in the fetal rib structure. The target fetal rib can be selected by the user or automatically determined by the system. Such as: the volume data of the fetal rib structure is displayed on a display, and the target fetal rib is determined according to rib identification manually input by a user or based on rib selection of the user on the fetal rib structure.

After the target fetal rib is determined, straightening the target fetal rib according to the three-dimensional volume data of the target fetal rib, so that the bent target fetal rib is straightened, and the straightened target rib is obtained. And after the three-dimensional volume data of the fetal spine are determined, straightening the fetal spine according to the three-dimensional volume data of the fetal spine, so that the bent fetal spine is straightened, and the straightened fetal spine is obtained. At this time, a second plane where the target fetal rib is located is determined in a manner that two straight lines determine one plane, wherein the second plane is perpendicular to the longitudinal axis of the straightened fetal spine and coplanar with the longitudinal axis of the straightened target rib. And after the second plane is determined, acquiring the three-dimensional volume data of the target fetal rib and the three-dimensional volume data of the fetal spine on the second plane to obtain the cross section of the fetal rib structure.

As used herein, the term "longitudinal axis" of a fetal rib, fetal spine, fetal lumbar spine or other tissue may refer to an axis along the length of the tissue, such as a central axis or other longitudinal axis.

In practical application, after the second plane is determined, the three-dimensional volume data of the straightened rib and the three-dimensional volume data of the straightened fetal spine of the target on the second plane are acquired, and the cross section of the straightened fetal rib structure is obtained.

It should be noted that the number of cross sections obtained corresponds to the number of fetal ribs in the target fetal rib, and different fetal ribs correspond to different cross sections.

In one embodiment, straightening the straightened object comprises: determining a longitudinal axis of the straightened subject; sampling the straightened object at equal intervals according to the longitudinal axis of the straightened object to obtain a tangent plane sequence at equal intervals; the straightening object comprises different fetal ribs or the fetal spine; and reconstructing the equally spaced tangent plane sequence along a straight line based on the equal spacing.

Straightening of the straightening object is divided into two parts of extraction of a longitudinal axis and straightening reconstruction. The extraction method of the longitudinal axis may be: a longitudinal axis extraction algorithm based on tracking, a multi-scale longitudinal axis extraction algorithm based on a model, a longitudinal axis extraction method based on morphology, a center line extraction method based on region growing, a method based on three-dimensional geometrical moments, a method for locating a center line by adopting machine learning, and the like.

For example, the tracking-based longitudinal axis extraction algorithm is a semi-automatic algorithm, a tangent plane perpendicular to the tracking direction is generated in the process of tracking the straightened object based on an initial key point and a termination point provided by a user, the central point of the straightened object in the tangent plane is accurately calculated by adopting a maximum likelihood value method and a centroid method, and the longitudinal axis of the straightened object is obtained by performing interpolation fitting on a central point sequence after the tracking process is finished.

For another example, when the model-based multi-scale longitudinal axis extraction algorithm is used for extracting the longitudinal axis of the straightening object, the local straightening object is approximated to be a tubular structure, the gravity center of the tubular structure obtained by calculating the geometric moment is used as the center of the local straightening object, the local straightening object is enhanced by analyzing the eigenvalue of the Hessian matrix corresponding to the multi-scale gaussian filtering lower body data, and the longitudinal axis direction of the straightening object is estimated according to the eigenvector corresponding to the minimum eigenvalue of the Hessian matrix.

After the longitudinal axis of the straightening object is determined, straightening reconstruction is carried out on the straightening object, the longitudinal axis of the rib is sampled at equal intervals to obtain equal-interval central points, and a tangent plane sequence which is at equal intervals and is perpendicular to the longitudinal axis direction of the straightening object is generated on the basis of the equal-interval central points. Stacking the obtained section sequences together, and performing three-dimensional reconstruction on the equally spaced section sequences to obtain a straightened object after straightening.

In an embodiment, the method further comprises: segmenting the three-dimensional volume data of the ribs and the spine of the fetus from the three-dimensional volume data of the first tissue to be detected; carrying out binarization processing on the three-dimensional volume data of the ribs and the three-dimensional volume data of the spines of the fetuses to obtain reconstructed three-dimensional volume data; rendering the reconstructed three-dimensional volume data to obtain the three-dimensional skeleton of the fetal rib structure.

After the three-dimensional volume data of the ribs of the fetus and the three-dimensional volume data of the spine of the fetus are identified, the three-dimensional volume data of the ribs of the fetus and the three-dimensional volume data of the spine of the fetus are subjected to binarization processing to obtain reconstructed three-dimensional volume data, such as: the local gray scale of the fetal rib structure is set to 1, and the volume data outside the fetal rib structure is set to 0. And rendering the reconstructed three-dimensional volume data by a volume rendering or surface rendering method to obtain a three-dimensional skeleton comprising the ribs and the spine of the fetus.

In practical applications, when the spine is identified, only the three-dimensional volume data of the spine connected with the ribs can be identified. Thereby outputting a rib structure including each of the fetal ribs and the vertebra.

According to the ultrasonic imaging method provided by the embodiment of the application, after the three-dimensional volume data of a fetus is acquired, the stereoscopic Virtual display (VR) image of the rib structure of the fetus is automatically imaged, and the rib of the fetus in the rib structure of the fetus is marked on the stereoscopic VR image, so that the workflow of the rib inspection of the fetus is greatly simplified. Further, the rib cage is three-dimensionally extracted, and the cross section and coronal plane of all ribs or a designated rib after being straightened are automatically imaged. The doctor is liberated from complicated manual operation, the dependency on the doctor technology is reduced, and the examination efficiency is improved.

An embodiment of the present application provides an ultrasound imaging method, which is applied to an ultrasound imaging apparatus shown in fig. 1, and as shown in fig. 3, the method includes:

s301, acquiring three-dimensional volume data of a first tissue to be detected;

the first tissue to be tested comprises: fetal rib structures and tissues outside of fetal rib structures such as: amniotic fluid region, placenta, uterine wall, etc.

Here, the tissue other than the fetal rib structure belongs to a low-echo region, and is low-grayscale data in the three-dimensional volume data, and the fetal rib structure belongs to a high-echo region, and is high-grayscale data in the three-dimensional volume data.

S302, based on a first rib detection model, respectively taking different fetal ribs or fetal spines as different identification objects, and identifying three-dimensional volume data of the fetal spines and three-dimensional volume data of the fetal ribs from the three-dimensional volume data of the first tissue to be detected;

the algorithm adopted by the first rib detection model can be a machine learning method, the first rib detection model takes three-dimensional volume data of each fetal rib and 3-volume data of a spine as training samples, the training samples are learned through the machine learning method, and the first rib detection model is trained through the training samples. The trained first rib detection model learns the volume data characteristics of the ribs and the spine of the fetus, wherein the volume data characteristics may include: principal Component Analysis (PCA), linear Discriminant Analysis (LDA), harr, texture, and the like.

When the first rib detection model receives the three-dimensional volume data of the first tissue to be detected, the three-dimensional volume data of the fetal rib and the three-dimensional volume data of the spine included in the received three-dimensional volume data are identified according to the learned volume data characteristics of the fetal rib and the learned volume data characteristics of the spine.

In the training sample, the fetal ribs or spines are used as targets, the targets are calibrated, and the category of each calibrated target is indicated. The calibration may be performed in a manner Of a Region Of Interest (ROI) frame including the target, or in a manner Of a Mask (Mask) for accurately segmenting the target.

The algorithm adopted by the first rib detection model can be an image segmentation algorithm, the three-dimensional volume data input into the first rib detection model is subjected to binarization segmentation, operations such as morphology, contour extraction and communication domain are carried out, a plurality of candidate regions are obtained, the probability that each candidate region is a fetal rib or spine is judged according to the volume data characteristics of each candidate region, the candidate region with the highest probability is selected as a region corresponding to the rib or spine, and the three-dimensional volume data of the selected region is the three-dimensional volume data of the fetal rib or spine.

In practical application, the image segmentation algorithm adopted by the first rib detection model may also be: level set (LevelSet), graph Cut (Graph Cut), snake, random walk (Random walk), active contour model algorithm, active shape model algorithm, active appearance model algorithm, and image segmentation algorithm in deep learning such as Full Convolution Network (FCN), UNet, etc.

The algorithm adopted by the first rib detection model can also be a template matching algorithm, and a template of the three-dimensional data of the ribs of the fetus or the three-dimensional data of the spine is established. The first rib detection model carries out binarization segmentation on input three-dimensional volume data, and carries out operations such as morphology, contour extraction, communication domain and the like to obtain a plurality of candidate regions, traverses all the candidate regions in the volume data according to an established template, determines the similarity of all the candidate regions and the template, selects the candidate region with the highest similarity as a target region corresponding to a fetal rib or a spine, and the three-dimensional volume data of the target region is the three-dimensional volume data of the fetal rib or the three-dimensional volume data of the spine.

It should be noted that, when the first rib model identifies different fetal ribs or fetal spines, the number of fetal ribs included in the first tissue to be tested may be directly counted.

S303, rendering the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs to obtain a three-dimensional ultrasonic image of a fetal rib structure;

after the three-dimensional volume data of the fetal rib and the three-dimensional volume data of the spine included in the first tissue to be detected are identified in S302, the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal rib are subjected to three-dimensional rendering, so as to obtain a three-dimensional ultrasonic image of the fetal rib structure.

When the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs are rendered in a three-dimensional mode, all data except the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs in the three-dimensional volume data of the first tissue to be measured are emptied, and the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs are rendered in a three-dimensional light perspective mode to form a three-dimensional ultrasonic image of a fetal rib structure.

S304, marking the fetal ribs in the three-dimensional ultrasonic image of the fetal rib structure;

and determining the position of the three-dimensional data of the fetal rib in the three-dimensional ultrasonic image of the fetal rib structure in the step S203 according to the position of the three-dimensional data of the fetal rib identified in the step S302 in the three-dimensional volume data of the first tissue to be detected, and marking the fetal rib in the fetal rib structure according to the determined position of the fetal rib to obtain the marked three-dimensional ultrasonic image of the fetal rib structure.

In marking the fetal ribs, the rib markers of one or more fetal ribs may be marked: such as: T1-T12, wherein T1 to T12 characterize the 1 st to 12 th fetal ribs, respectively.

And S305, outputting the marked three-dimensional ultrasonic image of the fetal rib structure.

According to the ultrasonic imaging method provided by the embodiment of the application, after the three-dimensional data of a fetus is obtained, the ribs and the spinal column of the fetus are taken as different identification objects, the ribs and the spinal column of the fetus are identified, the number of the ribs of the fetus included in a first tissue to be detected is automatically counted, and the three-dimensional VR image of the structure of the ribs of the fetus is automatically imaged, so that the working flow of the rib inspection of the fetus is greatly simplified, a doctor is liberated from complicated manual operation, the dependency on the technology of the doctor is reduced, and the misdiagnosis and the missed diagnosis rate are reduced.

An embodiment of the present application provides an ultrasound imaging method, which is applied to an ultrasound imaging apparatus shown in fig. 1, and as shown in fig. 4, the method includes:

s401, acquiring three-dimensional volume data of a first tissue to be detected;

Here, the tissue other than the fetal rib structure belongs to a hypoechoic region and is displayed in low gray in the ultrasound image, and the fetal rib structure belongs to a hyperechoic region and is displayed in high gray in the ultrasound image.

S402, identifying three-dimensional volume data of the fetal rib structure from the three-dimensional volume data of the first tissue to be detected;

when identifying the three-dimensional volume data of the fetal rib structure from the three-dimensional volume data of the first tissue to be detected, at least one of the following three structure identification modes can be included:

displaying a three-dimensional image corresponding to the three-dimensional volume data of the first tissue to be detected in a structure identification mode I; receiving a first input operation based on a three-dimensional image corresponding to the three-dimensional volume data of the first tissue to be detected; determining a mark point corresponding to the first input operation; and identifying the three-dimensional volume data of the fetal rib structure from the three-dimensional volume data of the first tissue to be detected according to the coordinates of the mark points.

After the three-dimensional volume data of the first tissue to be detected is acquired, a three-dimensional image corresponding to the three-dimensional volume data of the first tissue to be detected is displayed on a display. A user carries out first input operation in a fetal rib structure on a three-dimensional image corresponding to three-dimensional volume data of a first tissue to be detected displayed on a display through tools such as a trackball and a touch screen by selecting a marking point and drawing a marking line to inform ultrasonic imaging equipment of the position of the fetal rib structure in space. Here, the marker line is constituted by a plurality of continuous marker points. After the ultrasonic imaging equipment receives the first input operation, the position of the fetal rib structure in the three-dimensional volume data of the first tissue to be detected is determined through the coordinates of the corresponding mark points of the first input operation or the mark points forming the mark line.

Such as: the method comprises the steps that a user takes end points on each fetal rib or some discontinuous points on the boundary of the fetal rib structure, ultrasonic imaging equipment takes the points selected by the user as mark points, a center line of the fetal rib structure is roughly drawn or a boundary line of the fetal rib structure is drawn according to coordinates of the mark points, and the position of the fetal rib structure is determined according to the determined center line or boundary line, so that three-dimensional volume data of the fetal rib structure are obtained.

A structure identification mode II is used for displaying a three-dimensional image corresponding to the three-dimensional volume data of the first tissue to be detected; receiving a second input operation based on a three-dimensional image corresponding to the three-dimensional volume data of the first tissue to be detected; determining a first seed area corresponding to the second input operation; the first sub-region is positioned in a three-dimensional image region corresponding to the fetal rib structure; determining a first pixel characteristic of three-dimensional volume data of a first sub-region; and identifying the three-dimensional volume data of the fetal rib structure in the three-dimensional volume data of the first tissue to be detected according to the first pixel characteristic.

Here, after acquiring the three-dimensional volume data of the first tissue to be measured, a three-dimensional image corresponding to the three-dimensional volume data of the first tissue to be measured is displayed on the display. The user performs a second input operation on the three-dimensional image corresponding to the three-dimensional volume data of the first tissue to be measured displayed on the display based on a tool such as a trackball or a touch screen. After the ultrasonic imaging equipment receives a second input operation, determining that a region corresponding to the second input operation is a first sub-region, taking three-dimensional volume data of the first sub-region as prior data, acquiring pixel characteristics of edge gradient, gray scale and the like of the prior data as first pixel characteristics, and identifying the three-dimensional volume data of the fetal rib structure from the three-dimensional volume data of the first tissue to be detected according to the first pixel characteristics of the prior data.

The method for identifying the three-dimensional volume data of the fetal rib structure from the three-dimensional volume data of the first tissue to be detected according to the pixel characteristics of the prior data can comprise the following steps: one or more image processing methods of template matching, image feature extraction, edge extraction, and morphological operations may also include: the method for segmenting the image comprises one or more image segmentation methods of a graphcut algorithm, a grabcut algorithm, a level set method, an active contour model algorithm, an active shape model algorithm, a seed region growing method and a region segmentation merging method, and can also comprise the following steps: the method comprises one or more machine learning methods of a deep learning method, a support vector machine, adaboost and a random forest algorithm.

For example, when the three-dimensional volume data of the fetal rib structure is identified from the three-dimensional volume data of the first tissue to be detected through template matching, the priori data is used as a template, pixel characteristics such as edge gradient and gray level of the template are calculated, the three-dimensional volume data of the first tissue to be detected is traversed through the template, so that the optimal solution with the minimum difference with the pixel characteristics of the template is found, and identification of the fetal rib structure is achieved.

For another example: when the three-dimensional volume data of the fetal rib structure is identified from the three-dimensional volume data of the first tissue to be detected through a seed region growing method, a seed region is determined in the fetal rib structure region, pixels of the seed region are used as seed pixels, then according to the first pixel characteristics of the seed pixels, pixels meeting the first pixel characteristics are merged into the seed region from the three-dimensional volume data of the first tissue to be detected, newly added pixels are used as new seed pixels to be merged continuously until new pixels meeting conditions cannot be found, and finally the three-dimensional volume data of the fetal rib bone structure are identified.

Here, certain a priori data is acquired in the fetal rib structure data in advance through a certain user interaction workflow to serve as known information, and the difficulty of identification of the fetal rib structure is reduced through a first pixel feature of the known information.

Determining at least two first candidate regions from the three-dimensional volume data of the first tissue to be detected, and acquiring volume data characteristics of the three-dimensional volume data of each first candidate region; determining a first matching degree of each first candidate region and the fetal rib structure according to the volume data characteristics of each first candidate region; determining a first candidate region with the highest first matching degree as a target region corresponding to the fetal rib structure; and taking the three-dimensional volume data of the target area corresponding to the fetal rib structure as the three-dimensional volume data of the fetal rib structure.

Here, the three-dimensional volume data of the first tissue to be detected may be received by a rib structure detection model, the rib structure detection model determines at least two first candidate regions from the three-dimensional volume data of the first tissue to be detected, and obtains the volume data feature of the three-dimensional volume data of each first candidate region; determining a first matching degree of each first candidate region and the fetal rib structure by a rib structure detection model according to the volume data characteristics of each first candidate region; determining a first candidate region with the highest first matching degree as a target region corresponding to the fetal rib structure by a rib structure detection model; and the rib structure detection model takes the three-dimensional volume data of the target area corresponding to the fetal rib structure as the three-dimensional volume data of the fetal rib structure.

The algorithm adopted by the rib structure detection model can be a machine learning method, the rib structure detection model takes three-dimensional volume data of each fetal rib structure as a training sample, the training sample is learned through the machine learning method, and the rib structure detection model is trained through the training sample. The trained rib structure detection model learns the volume data characteristics of the rib structure of the fetus, wherein the volume data characteristics can include: PCA characteristic, LDA characteristic, harr characteristic, texture characteristic and the like.

When the rib structure detection model receives the three-dimensional volume data of the first tissue to be detected, the three-dimensional volume data of the fetal rib structure included in the received three-dimensional volume data is identified according to the learned volume data characteristics of the fetal rib structure.

In the training sample, the fetal rib structures are used as targets, the targets are calibrated, and the category of each calibrated target is indicated. The calibration may be performed in a manner Of a Region Of Interest (ROI) frame including the target, or in a manner Of a Mask (Mask) for accurately segmenting the target.

The rib structure detection model adopts an algorithm which can be an image segmentation algorithm, three-dimensional volume data input into the rib structure detection model are subjected to binarization segmentation, operations such as morphology, contour extraction, communication domain and the like are carried out, a plurality of first candidate regions are obtained, the probability that each first candidate region is a fetal rib structure is judged according to the volume data characteristics of each first candidate region, the first candidate region with the highest probability is selected as a target region corresponding to the fetal rib structure, and the three-dimensional volume data of the selected target region are the three-dimensional volume data of the fetal rib structure.

In practical application, the image segmentation algorithm adopted by the rib structure detection model may also be: the image segmentation method comprises one or more of level set (LevelSet), graph Cut (Graph Cut), snake, random walk (Random walk), active contour model algorithm, active shape model algorithm and active appearance model algorithm, and image segmentation algorithm in deep learning such as FCN and UNet.

The algorithm adopted by the rib structure detection model can also be a template matching algorithm, and a template of the three-dimensional volume data of the fetal rib structure is established. The rib structure detection model carries out binarization segmentation on input three-dimensional volume data, and carries out operations such as morphology, contour extraction, communication domain and the like to obtain a plurality of first candidate regions, traverses all the first candidate regions in the volume data according to an established template, determines the similarity of all the first candidate regions and the template, selects the first candidate region with the highest similarity as a target region corresponding to a fetal rib structure, and the three-dimensional volume data of the target region is the three-dimensional volume data of the fetal rib structure.

In the embodiment of the present application, a specific identification manner of the three-dimensional volume data for identifying the structure of the fetal rib from the three-dimensional volume data of the first tissue to be measured is not limited in any way.

S403, segmenting the three-dimensional volume data of the fetal rib structure from the three-dimensional volume data of the first tissue to be detected;

after the three-dimensional volume data of the fetal rib structure are identified, the three-dimensional volume data of the fetal rib structure are segmented from the three-dimensional volume data of the first tissue to be detected according to the position of the three-dimensional volume data of the fetal rib structure in the three-dimensional volume data of the first tissue to be detected.

In practical applications, S403 and S404 may be implemented simultaneously, and the three-dimensional volume data of the fetal rib structure is segmented from the three-dimensional volume data of the first tissue to be detected while identifying the three-dimensional volume data of the fetal rib structure.

S404, identifying three-dimensional volume data of a spine of the fetus and three-dimensional volume data of ribs of the fetus from the three-dimensional volume data of the rib structure of the fetus;

in S403, after the three-dimensional volume data of the fetal rib structure is segmented from the three-dimensional volume data of the first tissue to be measured, when the three-dimensional volume data of the fetal rib and the three-dimensional volume data of the spine are identified from the three-dimensional volume data of the fetal rib structure, at least one of the following two rib identification modes may be included:

a first rib identification mode, based on the shape characteristics of the spine of the fetus and the shape characteristics of the ribs of the fetus, locating the included spine and the ribs of the fetus from the three-dimensional data of the structure of the ribs of the fetus; displaying a three-dimensional image corresponding to the three-dimensional volume data of the fetal rib structure; receiving a third input operation based on a three-dimensional image corresponding to the three-dimensional volume data of the fetal rib structure; according to the third input operation, determining a reference fetal rib and a rib mark corresponding to the reference fetal rib; the reference fetal rib is at least one fetal rib in the fetal rib structure; and identifying the three-dimensional volume data of the ribs of the fetus based on the reference fetal ribs and the rib marks corresponding to the reference fetal ribs.

After the three-dimensional volume data of the fetal rib structure is determined, a three-dimensional image corresponding to the three-dimensional volume data of the fetal rib structure is displayed on a display. The user performs a third input operation based on a three-dimensional image corresponding to the three-dimensional volume data of the rib structure displayed on the display through a trackball, a touch screen, or the like, to identify a portion of the fetal ribs in the fetal rib structure and a rib identification of the identified fetal rib structure.

After the ultrasonic imaging device receives the third input operation, determining the fetal ribs identified by the user through the third input operation, taking the fetal ribs identified by the user as reference fetal ribs, taking the rib identifications input by the user as the rib identifications of the reference fetal ribs, and determining each fetal rib in the fetal rib structure and the rib identification corresponding to each fetal rib according to the reference fetal ribs. Here, the reference fetal rib may include one or more fetal ribs.

When the ultrasonic imaging device determines each fetal rib in the fetal rib structure and the rib mark corresponding to each fetal rib according to the reference fetal rib, the number of the fetal ribs included in the fetal rib structure and the number of the included fetal ribs can be determined according to one or more of a gray histogram projection method, a contour extraction algorithm, an edge extraction algorithm, a communication domain method, a blob detection algorithm, a template matching algorithm, an image feature extraction algorithm and a morphological operation algorithm.

Such as: a user selects T3 and T9 ribs on a three-dimensional image corresponding to a fetal rib structure through a trackball, a touch screen and other tools, one point is arranged on the T3 and T9 ribs or the central lines of the T3 and T9 ribs are respectively drawn, namely the two ribs are marked as reference fetal ribs. Projecting the coronal plane of the three-dimensional volume data of the fetal rib structure along the Y axis by a histogram gray projection method to obtain a pixel statistical map, and calculating the number of pixel wave crests and the positions of the peak values of the pixel statistical map to obtain the number and the positions of the fetal ribs. And determining the fetal rib above the T2 rib as T1, determining the fetal rib below the T2 rib as T3, and determining T4-T12 in the same way. Thereby, the fetal ribs in the fetal rib structure and the rib marks of the fetal ribs are determined according to the sequencing order and the spatial position relation of the positions of the reference fetal ribs and the identified fetal ribs.

And a second rib identification mode, based on a second rib detection model, respectively taking different fetal ribs or fetal spines as different identification objects, and identifying the three-dimensional volume data of the fetal spines and the three-dimensional volume data of the fetal ribs from the three-dimensional volume data of the fetal rib structures.

The algorithm adopted by the second rib detection model can be a machine learning method, the second rib detection model takes the three-dimensional volume data of each fetal rib and the three-dimensional volume data of the spine as training samples, the training samples are learned through the machine learning method, and the second rib detection model is trained through the training samples. The trained second rib detection model learns the volume data characteristics of the ribs and the spine of the fetus, wherein the volume data characteristics may include: PCA characteristic, LDA characteristic, harr characteristic, texture characteristic and the like.

When the second rib detection model receives the three-dimensional volume data of the fetal rib structure, the three-dimensional volume data of the fetal rib and the three-dimensional volume data of the spine included in the received three-dimensional volume data are identified according to the learned volume data characteristics of the fetal rib and the learned volume data characteristics of the spine.

In the training sample, the fetal ribs or spines are used as targets, the targets are calibrated, and the category of each calibrated target is indicated. The calibration can be performed in a manner of an ROI frame including the target, or in a manner of a Mask (Mask) for accurately segmenting the target.

The algorithm adopted by the second rib detection model can be an image segmentation algorithm, the three-dimensional volume data input into the second rib detection model is subjected to binarization segmentation, operations such as morphology, contour extraction and communication domain are carried out, a plurality of candidate regions are obtained, the probability that each candidate region is a fetal rib or spine is judged according to the volume data characteristics of each candidate region, the candidate region with the highest probability is selected as a region corresponding to the rib or spine, and the three-dimensional volume data of the selected region is the three-dimensional volume data of the fetal rib or spine.

In practical application, the image segmentation algorithm adopted by the second rib detection model may also be: level set (LevelSet), graph Cut (Graph Cut), snake, random walk (Random walk), active contour model algorithm, active shape model algorithm, active appearance model algorithm, and image segmentation algorithm in deep learning of Full Convolution Network (FCN), UNet, etc.

The algorithm adopted by the second rib detection model can also be a template matching algorithm, and a template of the three-dimensional data of the ribs of the fetus or the three-dimensional data of the spine is established. The second rib detection model carries out binarization segmentation on the input three-dimensional volume data, obtains a plurality of candidate regions after carrying out operations such as morphology, contour extraction, communication domain and the like, traverses all the candidate regions in the volume data according to the established template, determines the similarity of all the candidate regions and the template, selects the candidate region with the highest similarity as a target region corresponding to a fetal rib or a spine, and the three-dimensional volume data of the target region is the three-dimensional volume data of the fetal rib or the three-dimensional volume data of the spine.

The algorithm adopted by the second rib detection model can also be as follows: one or more methods such as image edge extraction, histogram image gray projection statistics, image contour extraction, morphology processing, threshold segmentation, block mass detection and the like are used for directly calculating the number of fetal ribs in the three-dimensional volume data of the fetal rib structure.

It should be noted that, when the second rib detection model identifies different fetal ribs or fetal spines, the number of fetal ribs included in the fetal rib structure can be directly counted.

S405, rendering the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs to obtain a three-dimensional ultrasonic image of a fetal rib structure;

after the three-dimensional volume data of the fetal rib and the three-dimensional volume data of the spine included in the first tissue to be measured are identified in S404, the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal rib are subjected to three-dimensional rendering to obtain a three-dimensional ultrasonic image of the fetal rib structure.

S406, marking the fetal ribs in the three-dimensional ultrasonic image of the fetal rib structure;

and determining the position of the three-dimensional data of the fetal rib in the three-dimensional ultrasonic image of the fetal rib structure in the S203 according to the position of the three-dimensional data of the fetal rib identified in the S404 in the three-dimensional volume data of the first tissue to be detected, and marking the fetal rib in the fetal rib structure according to the determined position of the fetal rib to obtain the marked three-dimensional ultrasonic image of the fetal rib structure.

In marking the fetal ribs, the rib identification of one or more fetal ribs may be marked: such as: T1-T12, wherein T1 to T12 characterize the 1 st to 12 th fetal rib, respectively.

And S407, outputting the marked three-dimensional ultrasonic image of the fetal rib structure.

According to the ultrasonic imaging method provided by the embodiment of the application, after the three-dimensional volume data of a fetus is acquired, the three-dimensional volume data of the rib structure of the fetus is automatically or semi-automatically identified, positioned and segmented, the three-dimensional VR image of the rib structure of the fetus is automatically imaged, the number of the ribs of the fetus is automatically or semi-automatically counted, and the ribs of the fetus in the rib structure of the fetus are marked on the three-dimensional VR image. Further, the rib cage is three-dimensionally extracted, and the cross section and coronal plane of all ribs or designated ribs after being straightened are automatically imaged. Thereby greatly simplifying the workflow of fetal rib examination, freeing doctors from complicated manual operation, reducing the dependence on the doctor technology and improving the examination efficiency; and the stability of the statistical result of the number of ribs and the imaging quality are in a better state compared with manual operation, so that misdiagnosis and missed diagnosis rate are reduced.

An embodiment of the present application provides an ultrasound imaging method, which is applied to an ultrasound imaging apparatus shown in fig. 1, and as shown in fig. 5, the method includes:

s501, acquiring three-dimensional volume data of a second tissue to be detected;

the second tissue under test includes: fetal spine and tissues other than fetal spine, such as: amniotic fluid region, placenta, uterine wall, etc.

The doctor can scan the pregnant woman through the probe to obtain the three-dimensional data of the second tissue to be detected.

S502, identifying three-dimensional volume data of spinal cord cones and three-dimensional volume data of lumbar vertebrae from the three-dimensional volume data of the second tissue to be detected;

after obtaining the three-dimensional volume data of the second tissue to be measured based on S201, the three-dimensional volume data of the spinal cone and the three-dimensional volume data of the lumbar vertebra are identified from the obtained three-dimensional volume data.

The spinal cone can be identified from the three-dimensional volume data of the second tissue to be detected through a spinal cone detection model.

In an embodiment, the identifying three-dimensional volume data of the spinal cone from the three-dimensional volume data of the second tissue to be measured includes: determining at least two second candidate regions from the three-dimensional volume data of the second tissue to be detected, and acquiring the volume data characteristic of the three-dimensional volume data of each second candidate region; determining a second matching degree of each second candidate region and the fetal spine according to the volume data characteristics of each second candidate region; determining a second candidate region with the highest second matching degree as a target region corresponding to the spinal cone; and taking the three-dimensional volume data of the target region corresponding to the spinal cone as the three-dimensional volume data of the spinal cone.

Here, the spinal cone detection model determines at least two second candidate regions from the three-dimensional volume data of the second tissue to be detected, and obtains volume data characteristics of the three-dimensional volume data of each second candidate region; determining a second matching degree of each second candidate region and the fetal spine by the spinal cord cone detection model according to the volume data characteristics of each second candidate region; determining a second candidate region with the highest second matching degree as a target region corresponding to the spinal cone by the spinal cone detection model; and taking the three-dimensional volume data of the target area corresponding to the spinal cone as the three-dimensional volume data of the spinal cone by the spinal cone detection model.

The spinal cone detection model adopts an algorithm which can be a machine learning method, the spinal cone detection model takes three-dimensional volume data of a spinal cone as a training sample, the training sample is learned through the machine learning method, and the spinal cone detection model is trained through the training sample. The trained spinal cone detection model learns the volume data characteristics of the spinal cone, wherein the volume data characteristics can include: PCA characteristic, LDA characteristic, harr characteristic, texture characteristic and the like.

And when the spinal cord cone detection model receives the three-dimensional volume data of the second tissue to be detected, identifying the three-dimensional volume data of the included spinal cord cone from the received three-dimensional volume data according to the learned volume data characteristics of the spinal cord cone.

In the training sample, the spinal cord cone is used as a target, and the target is calibrated. The calibration can be performed in a manner of an ROI frame including the target, or in a manner of a Mask (Mask) for accurately segmenting the target.

The spinal cord cone detection model adopts an algorithm which can be an image segmentation algorithm, three-dimensional volume data input into the spinal cord cone detection model are subjected to binary segmentation, operations such as morphology, contour extraction and communication domain are carried out, a plurality of second candidate regions are obtained, the probability that each second candidate region is a spinal cord cone is judged according to the volume data characteristics of each second candidate region, the first candidate region with the highest probability is selected as a target region corresponding to the spinal cord cone, and the three-dimensional volume data of the selected target region are the three-dimensional volume data of the spinal cord cone.

In practical application, the image segmentation algorithm adopted by the spinal cone detection model may also be: the image segmentation method comprises one or more of level set (LevelSet), graph Cut (Graph Cut), snake, random walk (Random walk), active contour model algorithm, active shape model algorithm and active appearance model algorithm, and image segmentation algorithm in deep learning such as FCN and UNet.

The algorithm adopted by the spinal cone detection model can also be a template matching algorithm, and a template of three-dimensional volume data of the spinal cone is established. The spinal cord cone detection model carries out binarization segmentation on input three-dimensional volume data, and carries out operations such as morphology, contour extraction, communication domain and the like to obtain a plurality of second candidate regions, traverses all the second candidate regions in the volume data according to an established template, determines the similarity between all the second candidate regions and the template, selects the second candidate region with the highest similarity as a target region corresponding to a spinal cord cone, and the three-dimensional volume data of the target region is the three-dimensional volume data of a fetal spinal cord cone.

It should be noted that, in the embodiment of the present application, a specific identification manner of identifying the three-dimensional volume data of the spinal cone from the three-dimensional volume data of the second tissue to be examined is not limited at all.

The method for identifying the three-dimensional volume data of the lumbar vertebra at least comprises one of the following two identification modes:

the identification method comprises the steps of firstly, taking the whole fetal spine as an identification object to identify the three-dimensional volume data of the fetal spine from the three-dimensional volume data of the second tissue to be detected, and then identifying the three-dimensional volume data of the lumbar spine from the three-dimensional volume data of the fetal spine.

And in the second identification mode, the lumbar vertebrae are directly used as identification objects to identify the lumbar vertebrae from the three-dimensional volume data of the second tissue to be detected.

In the first identification mode, identifying three-dimensional volume data of a fetal spine from the three-dimensional volume data of the second tissue to be detected; segmenting the three-dimensional volume data of the fetal spine from the three-dimensional volume data of the second tissue to be detected; and identifying the three-dimensional volume data of the lumbar vertebra from the three-dimensional volume data of the vertebral column of the fetus.

In the second recognition mode, the lumbar vertebrae are recognized as the recognition target based on the first lumbar vertebrae detection model, and the three-dimensional volume data of the lumbar vertebrae are recognized from the three-dimensional volume data of the second tissue to be measured.

In an embodiment, the identifying the three-dimensional volume data of the lumbar vertebra from the three-dimensional volume data of the second tissue to be measured includes: determining at least two third candidate regions from the three-dimensional volume data of the second tissue to be detected, and acquiring the volume data characteristics of the three-dimensional volume data of each third candidate region; determining a third matching degree of each third candidate region and lumbar vertebrae in the spinal column of the fetus according to the volume data characteristics of each third candidate region; determining a third candidate region with the highest third matching degree as a target region corresponding to the lumbar vertebra; and taking the three-dimensional volume data of the target area corresponding to the lumbar vertebra as the three-dimensional volume data of the lumbar vertebra.

Here, the first lumbar detection model determines at least two third candidate regions from the three-dimensional volume data of the second tissue to be detected, and obtains volume data characteristics of the three-dimensional volume data of each third candidate region; the first lumbar vertebra detection model determines a third matching degree of each third candidate region and lumbar vertebra in the spinal column of the fetus according to the volume data characteristics of each third candidate region; the first lumbar vertebra detection model determines a third candidate region with the highest third matching degree as a target region corresponding to the lumbar vertebra; the first lumbar vertebra detection model takes the three-dimensional volume data of the target area corresponding to the lumbar vertebra as the three-dimensional volume data of the lumbar vertebra.

The algorithm adopted by the first lumbar vertebra detection model can be a machine learning method, the first lumbar vertebra detection model takes the three-dimensional data of the lumbar vertebra as a training sample, the training sample is learned through the machine learning method, and the first lumbar vertebra detection model is trained through the training sample. The first lumbar vertebrae detection model through training learns the volume data characteristic of lumbar vertebrae, and wherein, the volume data characteristic can include: PCA characteristic, LDA characteristic, harr characteristic, texture characteristic and the like.

And when the first lumbar vertebra detection model receives the three-dimensional volume data of the second tissue to be detected, recognizing the included three-dimensional volume data of the lumbar vertebra from the received three-dimensional volume data according to the learned volume data characteristics of the lumbar vertebra.

S503, rendering the three-dimensional data of the spinal cord cone and the three-dimensional data of the lumbar vertebra to obtain a three-dimensional ultrasonic image of a vertebra structure;

after the three-dimensional volume data of the spinal cord cone and the three-dimensional volume data of the lumbar vertebra in the second tissue to be detected are identified in S502, three-dimensional rendering is performed on the three-dimensional volume data of the spinal cord cone and the three-dimensional volume data of the lumbar vertebra, and a three-dimensional ultrasonic image of the vertebra structure is obtained.

When the three-dimensional volume data of the spinal cord cone and the lumbar vertebra are rendered in a three-dimensional mode, all data except the three-dimensional volume data of the spinal cord cone and the three-dimensional volume data of the lumbar vertebra in the three-dimensional volume data of the second tissue to be detected are emptied, and three-dimensional light perspective rendering is performed on the three-dimensional volume data of the spinal cord and the three-dimensional volume data of the lumbar vertebra to form a three-dimensional ultrasonic image of a vertebral structure.

S504, marking the spinal cord cone in the three-dimensional ultrasonic image of the vertebra structure;

and determining the position of the three-dimensional data of the spinal cone in the three-dimensional ultrasonic image of the fetal rib bone structure in the step S203 according to the position of the three-dimensional data of the spinal cone identified in the step S502 in the three-dimensional volume data of the second tissue to be detected, and marking the spinal cone according to the determined position of the spinal cone to obtain a three-dimensional ultrasonic image of the marked vertebral structure.

Here, the three-dimensional ultrasound image of the displayed vertebral structure may include only the spinal conies and the lumbar vertebrae, or may include vertebrae such as vertebral bones other than the lumbar vertebrae.

In practical application, the distance between the spinal cord cone and the lumbar vertebra is detected, and the detected distance is displayed on a display screen. And comparing the detected distance with the set distance range, and sending out prompt information to prompt the user that the current position of the spinal cord cone is abnormal if the detected distance is not matched with the set distance range.

And S505, outputting the marked three-dimensional ultrasonic image of the vertebral structure.

At this time, the contents displayed on the display of the ultrasound imaging apparatus are: a three-dimensional ultrasound image of a vertebral structure marked with the spinal conus enables a user to visually view the location of the spinal conus of a fetus and determine the distance between the spinal conus and the lumbar spine.

In one embodiment, marking the spinal cone in a three-dimensional ultrasound image of the vertebral structure comprises: determining three-dimensional coordinates of the conical end of the spinal cord; determining a longitudinal axis of the lumbar spine; determining a reference line or a third plane through the spinal conical end and perpendicular to the longitudinal axis of the lumbar spine from the three-dimensional coordinates of the spinal conical end and the fetal spine longitudinal axis; mapping the reference line or the third plane in a three-dimensional ultrasound image of the vertebral structure.

The three-dimensional coordinates of the spinal conic tip in the three-dimensional volume data are calculated from the two-dimensional coordinates of the spinal conic tip in the sagittal plane and the specific position of the sagittal plane in the three-dimensional volume data, and a plane or a straight line passing through the end point of the spinal conic tip and perpendicular to the lumbar vertebrae is calculated from the three-dimensional coordinates of the spinal conic tip and the longitudinal axis of the lumbar vertebrae in the three-dimensional volume data, and the plane or the straight line is referred to as a third plane. And mapping the third plane to a stereo VR diagram of the fetal spine, so that the relative position of the conical end of the spinal cord relative to the lumbar vertebrae of the fetus can be visually represented, and the result is calculated and marked on the VR diagram. Wherein the longitudinal axis of the lumbar spine in the three-dimensional volumetric data characterizes the pose of the fetal spine.

According to the ultrasonic imaging method provided by the embodiment of the application, after the three-dimensional volume data of a fetus is obtained, the three-dimensional VR image of the spinal column of the fetus is automatically imaged, and the spinal cone is marked on the three-dimensional VR image, so that the workflow of spinal cone examination is greatly simplified. Further, the specific location of the spinal conus in the lumbar spine is marked on the stereo VR diagram. The doctor is liberated from complicated manual operation, the dependency on the doctor technology is reduced, and the examination efficiency is improved.

An embodiment of the present application provides an ultrasound imaging method, which is applied to an ultrasound imaging apparatus shown in fig. 1, and as shown in fig. 6, the method includes:

s601, acquiring three-dimensional volume data of a second tissue to be detected;

The doctor can scan the pregnant woman through the probe to acquire the three-dimensional data of the second tissue to be detected.

S602, identifying three-dimensional volume data of a spinal cord cone from the three-dimensional volume data of the second tissue to be detected;

after obtaining the three-dimensional volume data of the second tissue to be measured based on S601, the three-dimensional volume data of the spinal cone is identified from the obtained three-dimensional volume data.

The spinal cone can be identified from the three-dimensional volume data of the second tissue to be detected through a spinal cone detection model. Determining at least two second candidate regions from the three-dimensional volume data of the second tissue to be detected by the spinal cone detection model, and acquiring the volume data characteristics of the three-dimensional volume data of each second candidate region; determining a second matching degree of each second candidate region and the fetal spine by the spinal cord cone detection model according to the volume data characteristics of each second candidate region; determining a second candidate region with the highest second matching degree as a target region corresponding to the spinal cone by the spinal cone detection model; and the spinal cord cone detection model takes the three-dimensional volume data of the target area corresponding to the spinal cord cone as the three-dimensional volume data of the spinal cord cone.

S603, identifying the three-dimensional volume data of the vertebral column of the fetus from the three-dimensional volume data of the second tissue to be detected;

after obtaining the three-dimensional volume data of the second tissue to be measured based on S601, the three-dimensional volume data of the fetal spine is identified from the obtained three-dimensional volume data.

When the three-dimensional volume data of the fetal spine are identified from the three-dimensional volume data of the second tissue to be detected, at least one of the following three spine identification modes can be included:

the spine identification mode is I, and a three-dimensional image corresponding to the three-dimensional volume data of the second tissue to be detected is displayed; receiving a fourth input operation based on the three-dimensional image corresponding to the three-dimensional volume data of the second tissue to be detected; determining a mark point corresponding to the fourth input operation; and identifying the three-dimensional volume data of the spine from the three-dimensional volume data of the second tissue to be detected according to the coordinates of the mark point corresponding to the fourth input operation.

And after the three-dimensional volume data of the second tissue to be detected is acquired, displaying a three-dimensional image corresponding to the three-dimensional volume data of the first tissue to be detected on a display. And the user performs a fourth input operation in the fetus spine on the three-dimensional image corresponding to the three-dimensional volume data of the second tissue to be detected displayed on the display by means of a trackball, a touch screen and other tools through selecting a mark point, drawing a mark line and other methods to inform the ultrasonic imaging equipment of the position of the fetus spine in the space. Here, the marker line is constituted by a plurality of continuous marker points. And after the ultrasonic imaging equipment receives the fourth input operation, determining the position of the fetal spine in the three-dimensional data of the second tissue to be detected according to the coordinates of the corresponding mark points of the fourth input operation or the mark points forming the mark line.

Such as: the method comprises the steps that a user intermittently takes some points at the end point of each vertebra of a fetal spine or on the boundary of the fetal spine, an ultrasonic imaging device takes the points selected by the user as mark points, the midline of the fetal spine or the boundary line of the fetal spine is roughly drawn according to the coordinates of the mark points, and the position of the fetal spine is determined according to the determined midline or the boundary line, so that the three-dimensional data of the fetal spine is obtained.

A spine recognition mode II is adopted, and a three-dimensional image corresponding to the three-dimensional volume data of the second tissue to be detected is displayed; receiving a fifth input operation based on a three-dimensional image corresponding to the three-dimensional volume data of the second tissue to be detected; determining a second seed area corresponding to the fifth input operation; the second seed region is positioned in the three-dimensional image region corresponding to the spine; determining a second pixel characteristic of the three-dimensional volume data of the second seed region; and identifying the three-dimensional volume data of the spine in the three-dimensional volume data of the second tissue to be detected according to the second pixel characteristic.

Here, after acquiring the three-dimensional volume data of the second tissue to be examined, a three-dimensional image corresponding to the three-dimensional volume data of the second tissue to be examined is displayed on the display. The user performs a fifth input operation on the three-dimensional image corresponding to the three-dimensional volume data of the second tissue to be measured displayed on the display based on a tool such as a trackball or a touch screen. After the ultrasonic imaging equipment receives the fifth input operation, determining that the region corresponding to the fifth input operation is a second seed region, taking the three-dimensional volume data of the second seed region as prior data, acquiring the pixel characteristics of edge gradient, gray level and the like of the prior data as second pixel characteristics, and identifying the three-dimensional volume data of the vertebral column of the fetus from the three-dimensional volume data of the second tissue to be detected according to the second pixel characteristics of the prior data.

The method for identifying the three-dimensional volume data of the fetal spine from the three-dimensional volume data of the second tissue to be tested according to the pixel characteristics of the prior data can comprise the following steps: one or more image processing methods of template matching, image feature extraction, edge extraction, and morphological operations may also include: the method for segmenting the image comprises one or more image segmentation methods of a graphcut algorithm, a grabcut algorithm, a level set method, an active contour model algorithm, an active shape model algorithm, a seed region growing method and a region segmentation merging method, and can also comprise the following steps: the method comprises one or more machine learning methods of a deep learning method, a support vector machine, adaboost and a random forest algorithm.

For example, when the three-dimensional volume data of the fetal spine is identified from the three-dimensional volume data of the second tissue to be detected through template matching, the priori data is used as the template, the pixel characteristics such as edge gradient and gray scale of the template are calculated, the three-dimensional volume data of the second tissue to be detected is traversed through the template, so as to find the optimal solution with the minimum difference with the pixel characteristics of the template, and the identification of the fetal rib structure is realized.

For another example: when the fetal spine three-dimensional volume data is identified from the fetal spine three-dimensional volume data of the second tissue to be detected through a seed region growing method, determining a seed region in the fetal spine region, taking pixels of the seed region as seed pixels, combining pixels meeting second pixel characteristics into the seed region from the fetal spine three-dimensional volume data of the second tissue to be detected according to second pixel characteristics of the seed pixels, continuously combining newly added pixels as new seed pixels until new pixels meeting conditions cannot be found, and finally identifying the fetal spine three-dimensional volume data.

Here, certain a priori data is obtained in the three-dimensional volume data of the fetal spine as known information in advance through a certain user interactive operation workflow, and the difficulty of identification of the fetal spine is reduced through the pixel characteristics of the known information.

Determining at least two fourth candidate regions from the three-dimensional volume data of the second tissue to be detected, and acquiring the volume data characteristics of the three-dimensional volume data of each fourth candidate region; determining a fourth matching degree of each fourth candidate region and the vertebral column of the fetus according to the volume data characteristics of each fourth candidate region; determining a fourth candidate region with the highest fourth matching degree as a target region corresponding to the spinal column of the fetus; and taking the three-dimensional volume data of the target region corresponding to the fetal spine as the three-dimensional volume data of the fetal spine.

Here, the spine detection model may receive three-dimensional volume data of a second tissue to be detected, determine at least two fourth candidate regions from the three-dimensional volume data of the second tissue to be detected, and obtain volume data characteristics of the three-dimensional volume data of each fourth candidate region; determining a fourth matching degree of each fourth candidate region and the fetal spine by the spine detection model according to the volume data characteristics of each fourth candidate region; determining a fourth candidate region with the highest fourth matching degree as a target region corresponding to the fetal spine by the spine detection model; and the spine detection model takes the three-dimensional volume data of the target region corresponding to the fetal spine as the three-dimensional volume data of the fetal spine.

The spine detection model adopts an algorithm which can be a machine learning method, the three-dimensional data of the fetal spine is used as a training sample in the spine detection model, the training sample is learned through the machine learning method, and the spine detection model is trained through the training sample. The trained spine detection model learns the volume data characteristics of the spine of the fetus, wherein the volume data characteristics may include: PCA characteristic, LDA characteristic, harr characteristic, texture characteristic and the like.

And when the spine detection model receives the three-dimensional volume data of the second tissue to be detected, identifying the included three-dimensional volume data of the fetal spine from the received three-dimensional volume data according to the learned volume data characteristics of the fetal spine.

In the training sample, the target is calibrated by taking the vertebral column of the fetus as the target. The calibration can be performed in a manner of an ROI frame including the target, or in a manner of a Mask (Mask) for accurately segmenting the target.

The spine detection model adopts an algorithm which can be an image segmentation algorithm, three-dimensional volume data input into the spine detection model are subjected to binarization segmentation, operations such as morphology, contour extraction and communication domain are carried out, a plurality of fourth candidate regions are obtained, the probability that each fourth candidate region is a fetal spine is judged according to the volume data characteristics of each fourth candidate region, the fourth candidate region with the highest probability is selected as a target region corresponding to the fetal spine, and the three-dimensional volume data of the selected target region is the three-dimensional volume data of the fetal spine.

In practical application, the image segmentation algorithm adopted by the spine detection model may also be: the image segmentation method comprises one or more of level set (LevelSet), graph Cut (Graph Cut), snake, random walk (Random walk), active contour model algorithm, active shape model algorithm and active appearance model algorithm, and image segmentation algorithm in deep learning such as FCN and UNet.

The algorithm adopted by the spine detection model can also be a template matching algorithm, and a template of the three-dimensional volume data of the fetal spine is established. The spine detection model carries out binarization segmentation on input three-dimensional volume data, and carries out operations such as morphology, contour extraction and communication domain to obtain a plurality of fourth candidate regions, traverses all the fourth candidate regions in the volume data according to the established template, determines the similarity between all the fourth candidate regions and the template, namely fourth matching degree, selects the fourth candidate region with the highest similarity as a target region corresponding to a fetal spine, and the three-dimensional volume data of the target region is the three-dimensional volume data of the fetal spine.

It should be noted that, in the embodiment of the present application, a specific identification manner of identifying the three-dimensional volume data of the fetal spine from the three-dimensional volume data of the second tissue to be measured is not limited in any way.

S604, segmenting the three-dimensional volume data of the fetal spine from the three-dimensional volume data of the second tissue to be detected;

and after the three-dimensional volume data of the fetal spine are identified, segmenting the three-dimensional volume data of the fetal spine from the three-dimensional volume data of the second tissue to be detected according to the position of the three-dimensional volume data of the fetal spine in the three-dimensional volume data of the second tissue to be detected.

In practical applications, S603 and S604 may be implemented simultaneously, and the three-dimensional volume data of the fetal spine is segmented from the three-dimensional volume data of the second tissue to be measured while the three-dimensional volume data of the fetal spine is identified.

S605, identifying the three-dimensional volume data of the lumbar vertebra and the three-dimensional volume data of the lumbar vertebra from the three-dimensional volume data of the fetal spine;

in S604, after the three-dimensional volume data of the fetal spine is segmented from the three-dimensional volume data of the second tissue to be measured, when the three-dimensional volume data of the lumbar vertebra is identified from the three-dimensional volume data of the fetal spine, at least one of the following two lumbar vertebra identification methods may be included:

positioning vertebrae of the fetal spine from the three-dimensional volume data of the fetal spine based on shape features of the vertebrae in a lumbar vertebra recognition mode I; displaying a three-dimensional image corresponding to the three-dimensional volume data of the spinal column of the fetus; receiving a sixth input operation based on a three-dimensional image corresponding to the three-dimensional volume data of the spinal column of the fetus; determining a reference vertebra and a vertebra mark corresponding to the reference vertebra according to the sixth input operation; the reference vertebra is at least one vertebra in the fetal spine; and identifying three-dimensional volume data of the lumbar vertebra based on the reference vertebra and the vertebra identification corresponding to the reference vertebra.

After the three-dimensional volume data of the spine of the fetus is determined, a three-dimensional image corresponding to the three-dimensional volume data of the spine of the fetus is displayed on a display. The user performs a sixth input operation to identify a part of vertebrae in the fetal spine and the identified vertebrae identification of the fetal spine based on a three-dimensional image corresponding to the three-dimensional volume data of the spine displayed on the display through a trackball, a touch screen, or the like.

After the ultrasonic imaging device receives the sixth input operation, vertebrae identified by the user are determined through the sixth input operation, the vertebrae identified by the user are used as reference vertebrae, the vertebrae identification input by the user is used as the vertebrae identification of the reference vertebrae, and the vertebrae identification corresponding to lumbar vertebrae and lumbar vertebrae in the fetal spine is determined according to the reference vertebrae.

When the ultrasonic imaging device determines the lumbar vertebrae in the fetal spine and the rib marks corresponding to the lumbar vertebrae according to the reference vertebrae, the vertebrae included in the fetal spine can be determined according to one or more of a gray histogram projection method, a contour extraction algorithm, an edge extraction algorithm, a communication domain method, a mass detection algorithm, a template matching algorithm, an image feature extraction algorithm and a morphological operation algorithm, and the lumbar vertebrae are determined from the determined vertebrae included in the fetal spine.

And the second lumbar vertebra identification mode is based on the lumbar vertebra detection model aiming at the lumbar vertebra, the three-dimensional volume data of the lumbar vertebra are identified from the three-dimensional volume data of the fetal spine, and the lumbar vertebra detection model is obtained through sample volume data training of the lumbar vertebra.

The algorithm adopted by the lumbar vertebra detection model can be a machine learning method, the lumbar vertebra detection model takes the three-dimensional data of the lumbar vertebra as a training sample, the training sample is learned through the machine learning method, and the lumbar vertebra detection model is trained through the training sample. The volume data characteristic of lumbar vertebrae is learnt out to the lumbar vertebrae detection model through the training, and wherein, the volume data characteristic can include: PCA characteristic, LDA characteristic, harr characteristic, texture characteristic and the like.

When the lumbar vertebra detection model receives the three-dimensional volume data of the fetal spine structure, the three-dimensional volume data of the included lumbar vertebra is identified from the received three-dimensional volume data according to the learned volume data characteristics of the lumbar vertebra.

In the training sample, the lumbar vertebrae are used as targets, and the targets are calibrated. The calibration can be performed in a manner of an ROI frame including the target, or in a manner of a Mask (Mask) for accurately segmenting the target.

The algorithm adopted by the lumbar vertebra detection model can be an image segmentation algorithm, binarization segmentation is carried out on three-dimensional volume data input into the lumbar vertebra detection model, a plurality of candidate regions are obtained after operations such as morphology, contour extraction and communication domain are carried out, the probability that each candidate region is the lumbar vertebra is judged according to the volume data characteristics of each candidate region, the candidate region with the highest probability is selected as the region corresponding to the lumbar vertebra, and the three-dimensional volume data of the selected region is the three-dimensional volume data of the lumbar vertebra.

In practical application, the image segmentation algorithm adopted by the lumbar vertebra detection model may also be: level set (LevelSet), graph Cut (Graph Cut), snake, random walk (Random walk), active contour model algorithm, active shape model algorithm, active appearance model algorithm, and image segmentation algorithm in deep learning of Full Convolution Network (FCN), UNet, etc.

The algorithm adopted by the lumbar vertebra detection model can also be a template matching algorithm, and a template of the three-dimensional volume data of the lumbar vertebra is established. The method comprises the steps that a lumbar vertebra detection model conducts binarization segmentation on input three-dimensional volume data, a plurality of candidate regions are obtained after operations such as morphology, contour extraction and communication regions are conducted, all the candidate regions in the volume data are traversed according to an established template, the similarity of all the candidate regions and the template is determined, the candidate region with the highest similarity is selected as a target region corresponding to a fetal rib or a spine, and the three-dimensional volume data of the target region is the three-dimensional volume data of lumbar vertebrae.

The algorithm adopted by the lumbar vertebra detection model can also be as follows: one or more methods such as image edge extraction, histogram image gray projection statistics, image contour extraction, morphology processing, threshold segmentation, block mass detection and the like are used for directly calculating the number of fetal ribs in the three-dimensional volume data of the fetal rib structure.

In practical applications, the lumbar vertebra detection model and the first lumbar vertebra detection model may be a through-one detection model.

S605, rendering the three-dimensional data of the spinal cord cone and the three-dimensional data of the lumbar vertebra to obtain a three-dimensional ultrasonic image of a vertebra structure;

and after the three-dimensional volume data of the spinal cord cone in the second tissue to be detected is identified in the step S602, performing three-dimensional rendering on the three-dimensional volume data of the spinal cord cone and the three-dimensional volume data of the lumbar vertebra after the three-dimensional volume data of the lumbar vertebra in the second tissue to be detected is identified in the step S605, and obtaining a three-dimensional ultrasonic image of the vertebra structure.

When the three-dimensional volume data of the spinal cord cone and the lumbar vertebra are rendered in a three-dimensional mode, all data except the three-dimensional volume data of the spinal cord cone and the three-dimensional volume data of the lumbar vertebra in the three-dimensional volume data of the second tissue to be detected are emptied, and the three-dimensional volume data of the spinal cord and the three-dimensional volume data of the lumbar vertebra are rendered in a three-dimensional light perspective mode to form a three-dimensional ultrasonic image of a vertebral structure.

S606, marking the spinal cord cone in the three-dimensional ultrasonic image of the vertebra structure;

In practical application, the distance between the spinal cord cone and the lumbar vertebra is detected, and the detected distance is displayed on a display screen. And comparing the detected distance with the set distance range, and giving an alarm if the detected distance is not matched with the set distance range.

And S607, outputting the marked three-dimensional ultrasonic image of the vertebra structure.

According to the ultrasonic imaging method provided by the embodiment of the application, after the three-dimensional volume data of a fetus is obtained, the three-dimensional volume data of a spinal column of the fetus is automatically or semi-automatically identified, positioned and segmented, a three-dimensional VR diagram of a vertebral structure is automatically imaged, and a spinal cone is marked on the three-dimensional VR diagram. Thereby greatly simplifying the workflow of the spinal cord conicity examination, freeing doctors from complicated manual operation, reducing the dependence on the doctors' technology and improving the examination efficiency; and the stability of the spinal cord cone positioning result and the imaging quality are in a better state compared with manual operation, so that misdiagnosis and missed diagnosis rate are reduced.

In the related art, in the ultrasonic fetal rib examination and the spinal cord cone position examination, a fully manual mode is still adopted. For example: after acquiring the ultrasonic Volume data of the fetal rib, doctors often need to adjust the rotation and translation of the X, Y, Z axis to adjust the orientation of the data and manually adjust the size and position of the Volume of Interest VOI (Volume of Interest) so as to better observe the overall structure of the fetal rib structure or spine. For the fetal rib map shown in fig. 7, the number of fetal ribs or abnormal conditions in the region 701 corresponding to the rib structure needs to be counted manually to manually check whether the ribs are true or abnormal, and if the ribs a in the region 701 are abnormal, the positions of the ribs a need to be determined manually.

If it is desired to obtain a transverse section or a coronal plane of a certain fetal rib, all fetal ribs or vertebrae, as shown in fig. 8, it is necessary to manually draw an anatomical trajectory 801 on the fetal rib or vertebrae in fig. 8 (a) by using a Curved Multi-planar reconstruction (CMPR) function, and set parameters such as thickness, etc. to obtain a stretched transverse section or coronal plane 8 (B) of a specific rib or vertebra, so as to observe abnormal conditions of the diseased rib or vertebra. In the examination of the spinal cord conical position, a doctor needs to manually and continuously adjust the sagittal plane to the optimal position, namely, a section which subjectively considers that the spinal cord conical imaging effect is the best, for example, the spinal cord conical shown in fig. 9, and observes the position of the spinal cord conical on the lumbar relative position through human eyes, which is subjective and is easily influenced by the imaging quality and the observation angle to cause the inaccuracy of the result.

It can be seen that in the related art, each manual operation in the examination step of the fetal rib or spinal cord cone is relatively complicated, time-consuming and depends heavily on the skill and experience of the doctor, for example, in the step of operating CMPR, manual tracing requires high skill and patience, and more importantly, how well the doctor traces directly affects the quality of the transverse plane or coronal plane of the rib or vertebra, so that the imaging effect and quality are not consistent and stable, thereby affecting the diagnosis result.

The ultrasonic imaging method can effectively help doctors to perform disease auxiliary diagnosis, remarkably improves the working efficiency, and improves the quality of acquired key diagnosis data. As an example, an ultrasound imaging apparatus to which the ultrasound imaging method of the embodiment of the present application is applied may include, as shown in fig. 10: a volume data acquisition module 1001, a recognition module 1002, a statistics module 1003, an imaging module 1004, and a display module 1005. The modules shown in fig. 10 may be located in the processor 105 of the ultrasound imaging apparatus shown in fig. 1.

A volume data acquisition module 1001 for acquiring three-dimensional volume data of ultrasonic fetal ribs, fetal spine (also called spine) and spinal cord.

When the transmitting circuit 101 in fig. 1 transmits a group of pulses subjected to delay focusing to the probe 102, the probe 102 transmits ultrasonic waves to the tissue to be detected, receives ultrasonic echoes with tissue information reflected from the tissue to be detected after a certain time delay, and converts the ultrasonic echoes into electrical signals again. The receiving circuit 103 receives these electrical signals and sends these ultrasound echo signals to the beam forming module 104. The ultrasonic echo signal is focused, delayed, weighted and summed in the beam forming module 104, and then processed by the processor 105 to obtain the three-dimensional data including the fetal rib structure and the spine.

The identification module 1002 is used for identification, positioning and segmentation of an ultrasonic rib structure or spine, and identification, positioning and segmentation of a spinal cone.

The identification and positioning method of the fetal rib structure (also called as a rib structure) can be divided into manual, semi-automatic and automatic, and the identification mode is divided into two modes of identifying the whole structure of a rib framework and respectively identifying each rib (T1-T12) and a center line. The identification of fetal ribs is as follows:

1. manual acquisition of fetal rib structure position

The user can select mark points, draw mark lines and other methods on the rib structure in the volume data through a certain workflow by means of a trackball, a touch screen and other tools to inform the system of the position of the rib structure in space, for example, selecting the end point of each fetal rib or taking some points which are discontinuous on the boundary of the rib structure, drawing the middle line of the rib structure approximately or drawing the boundary line of the rib structure and the like.

2. Method for automatically identifying fetal rib structure

The method comprises the steps of learning characteristics or rules of fetal ribs and other non-rib structure tissues in a database by adopting a machine learning method, positioning and identifying the fetal rib structures in other body data according to the learned characteristics or rules, and constructing the database by using the body data of a plurality of fetal ribs and corresponding calibration results. The calibration result may be set according to actual task requirements, may be an ROI frame including the target, or may be a Mask (Mask) for accurately segmenting the target.

It should be noted that: if each rib or spine of the fetus is used as a different category to identify and locate the target rather than the structure formed by all ribs and spines as a whole target, the category of the rib or spine of each ROI box or Mask needs to be specified, that is, all the ribs and spines of the fetus need to be calibrated into different categories, which is converted into a multi-target identification and location problem. After the database is constructed, a machine learning algorithm is designed to learn the characteristics or rules of the fetal ribs or spine regions and the non-fetal ribs or non-spine regions in the database of the fetal ribs and the spine so as to realize the positioning and identification of the fetal ribs and the spine in the volume data.

Methods of identifying fetal rib structures include, but are not limited to, the following:

the method comprises the following steps: firstly, extracting the features of the area in the sliding window, wherein the extracted features can be PCA features, LDA features, harr features, texture features and the like, or extracting the features by adopting a deep neural network, then matching the extracted features with a database, classifying by using discriminators such as KNN, SVM, random forest, neural network and the like, and determining whether the current sliding window is the region of interest and acquiring the corresponding category of the region of interest.

And secondly, identifying by a border-Bounding-Box (Bounding-Box) method based on deep learning, wherein the common form is as follows: the constructed database is subjected to characteristic learning and parameter regression by stacking the base layer convolution layer and the full connection layer, the corresponding Bounding-Box of the region of interest can be directly regressed through the network for the input three-dimensional volume data, and the category of the organization structure in the region of interest can be obtained at the same time, and the algorithm adopted by the network model is R-CNN, fast-RCNN, SSD, YOLO and the like.

And a third method, an end-to-end semantic segmentation network method based on deep learning, wherein the method is similar to the second method, namely a Bounding-Box based on deep learning in structure, and is different in that a full connection layer is removed, an up-sampling or anti-convolution layer is added to enable the input size and the output size to be the same, so that the region of interest of the input image and the corresponding category of the region of interest are directly obtained, and common networks comprise FCN, U-Net, mask R-CNN and the like.

And a fourth method, adopting the first method, the second method or the third method to position the target, and then classifying and judging the target of the positioning result through a classifier. The class judgment method can be as follows: firstly, extracting the characteristics of a target ROI or Mask, wherein the extracted characteristics can be PCA characteristics, LDA characteristics, harr characteristics, texture characteristics and the like, or deep neural networks can be adopted for extracting the characteristics, then the extracted characteristics are matched with a database, and classification is carried out by discriminators such as KNN, SVM, random forest, neural networks and the like.

The method for identifying the fetal rib structure can be an image segmentation algorithm, and the fetal rib structure in the volume data can be accurately segmented. Here, the fetal rib structure is usually represented as a high-echo arc-band-shaped object with high gray value, and the rib structure can be segmented by an image segmentation algorithm.

For example, first, the volume data is divided into two, and then, a plurality of candidate regions are obtained after performing operations such as morphology, contour extraction, and communication domain, and then, for each candidate region, the probability that the region is a rib structure is judged according to the characteristics such as shape, length-width ratio, and the like, and the region with the highest probability is selected as the rib structure region. Other image segmentation algorithms, such as one or more of level set (LevelSet), graph Cut (Graph Cut), snake, random walk (Random walk), active contour model algorithms, active shape model algorithms, active appearance model algorithms, and some in-deep learning image segmentation algorithms, such as FCN, UNet, and so forth, may also be employed.

For another example, a template matching method may also be used to detect the fetal rib structure in the volume data, for example, the shape of the fetal rib structure is very specific, some data of the fetal rib structure may be collected in advance to establish a template, all possible regions in the volume data are traversed during detection, similarity matching is performed with the template, and the region with the highest similarity is selected as the target region.

3. The method for semi-automatically acquiring the position of the fetal rib comprises the following steps:

the method comprises the steps that a user uses tools such as a track ball and a touch screen, a certain priori data is obtained in the volume data of the fetal rib structure in advance through a certain user interactive operation workflow and serves as known information or a certain parameter or condition is preset to help reduce recognition difficulty and the like, and then one or more image processing methods of a template matching algorithm, an image feature extraction algorithm, an edge extraction algorithm and a morphological operation algorithm, one or more image segmentation algorithms of a graphcut algorithm, a grabcut algorithm, a level set method, an active contour model algorithm, an active shape model algorithm, a seed region growing method and a region segmentation merging method, and one or more machine learning methods of a deep learning method, a support vector machine, an adaboost and a random forest algorithm are combined to achieve recognition and determine the position of the fetal rib structure in the volume data.

For example, when the template matching method is adopted, a sample volume data needs to be obtained in advance, a trackball, a touch screen and the like can be used for taking out data containing a part or all of the fetal rib structure information to manufacture a template, for example, an image containing the fetal rib structure on one or more sections is intercepted by a rectangular frame to serve as the template, then information such as edge gradient and gray scale of the template is calculated, and then the template is used for traversing the whole volume data to search for the optimal solution with the minimum difference with the template information to realize identification of the fetal ribs.

When the graphicut algorithm is adopted to segment the fetal rib region, it is also necessary to first designate pixel points respectively representing a target region and a background region in the fetal rib structure region and the non-fetal rib structure region as seed points, for example, a user designates different seed points by respectively pointing points and drawing lines inside and outside the rib structure through a trackball; the grabcut algorithm needs to manually draw a frame to frame the segmented rib structure target region or draw lines with different colors respectively designating pixels of the target region and the background region in the rib structure region and the non-rib structure region respectively, so as to obtain a perfect segmentation result.

The segmentation methods such as the level set and the active contour model require user interaction to give an initial contour curve, and then curve evolution is carried out according to functional energy minimization to approach the boundary of the fetal rib, so that the segmentation of the fetal rib is realized; the seed region growing method also requires a user to provide a group of seed pixels representing different growing regions, then, the pixels which are in accordance with the field of the seed pixels are merged into the growing regions represented by the seed pixels, the newly added pixels are continuously merged as new seed pixels until new pixels which are in accordance with conditions cannot be found, and finally, the fetal rib structures are segmented. Or the detection range can be narrowed by the method that the user sets a rectangular box or appoints some mark points of the rib structure through the interactive workflow in advance, and the like, so that the identification difficulty is reduced.

By the above method, the position of the fetal rib structure in the volume data can be identified.

In practical applications, the above-described manner of identifying fetal rib structures is equally applicable to identification of the spine.

It should be noted that, when constructing the database for machine learning and the template for template matching, the sample may be an entire structure including the spine and all 12 ribs as a labeled sample or template, or the 12 ribs and the spine may be regarded as different classes of multi-target de-labeling and recognition or template matching.

Identification and location of the spinal conus requires automatic identification, segmentation and determination of the exact location of the spinal conus in the spinal volume data. The method comprises the following steps:

the method comprises the steps of learning characteristics or rules of a spinal cone and other non-spinal cone tissues in a database by adopting a machine learning method, positioning and identifying the spinal cone in other volume data according to the learned characteristics or rules, and firstly constructing the database by using multi-frame pictures for sampling a plurality of spinal data and corresponding calibration results. The calibration result can be set according to actual task requirements, and can be an ROI (region of interest) frame containing a target or a Mask (Mask) for accurately segmenting the target; after the database is built, a machine learning algorithm is designed to learn the characteristics or rules of the spinal cone database which can distinguish the spinal cone region from the non-spinal cone region, so as to realize the positioning and identification of the spinal cone in the volume data.

Common methods include, but are not limited to, the following:

the method I is based on a sliding window method: firstly, extracting the features of the area in the sliding window, wherein the extracted features can be PCA features, LDA features, harr features, texture features and the like, or deep neural networks can be adopted for extracting the features, then the extracted features are matched with a database, classifiers such as KNN, SVM, random forest, neural networks and the like are used for classifying, and whether the current sliding window is the region of interest or not is determined, and the corresponding category is obtained at the same time.

The second method is detection and identification based on a Bounding-Box method of deep learning, and the common forms are as follows: the constructed database is subjected to characteristic learning and parameter regression by stacking the base layer convolution layer and the full connection layer, the corresponding Bounding-Box of the region of interest can be directly regressed through a network for the input three-dimensional volume data, and the category of the organization structure in the region of interest can be obtained at the same time. The network model can be implemented by R-CNN, fast-RCNN, SSD, YOLO, etc.

And the method is similar to the structure of a second Bounding-Box based on deep learning, and is different in that a full connection layer is removed, and an up-sampling or anti-convolution layer is added to enable the input and output sizes to be the same, so that the region of interest of an input image and the corresponding category of the region of interest are directly obtained, and common networks comprise FCN, U-Net, mask R-CNN and the like.

And fourthly, positioning the target by only adopting the first method, the second method or the third method, and classifying and judging the target of the positioning result by the classifier. The classification judgment method can be as follows: firstly, extracting the characteristics of a target ROI or Mask, wherein the extracted characteristics can be PCA characteristics, LDA characteristics, harr characteristics, texture characteristics and the like, or extracting the characteristics by adopting a deep neural network, then matching the extracted characteristics with a database, and classifying by using discriminators such as KNN, SVM, random forest, neural network and the like.

The method for identifying the spinal cord cones can also be an image segmentation algorithm, and the spinal cord cones in the volume data can be accurately segmented. Here, the spinal cord cone usually represents a hyperechoic arc-band-shaped object with high gray values, and can be segmented by an image segmentation algorithm.

For example, firstly, the volume data is subjected to binarization segmentation, and a plurality of candidate regions are obtained after operations such as necessary morphology, contour extraction, communication domain and the like are performed, then, the probability that each candidate region is a spinal cone is judged according to characteristics such as shape, length-width ratio and the like, and a region with the highest probability is selected as a spinal cone region. Other image segmentation algorithms, such as one or more of level set (LevelSet), graph Cut (Graph Cut), snake, random walk (Random walk), active contour model algorithms, active shape model algorithms, active appearance model algorithms, and some in-deep learning image segmentation algorithms, such as FCN, UNet, and so forth, may also be employed.

For another example, a template matching method may also be used to detect a spinal cone in the volume data, for example, the fetal spinal cone has a more specific spinal cone shape, some data of the fetal spinal cone may be collected in advance to establish a template, all possible regions in the volume data are traversed during detection, similarity matching is performed with the template, and the region with the highest similarity is selected as the target region.

The accurate position of the spinal cord cone of the fetus in the volume data can be automatically positioned by the method.

And a statistic module 1003 used for semi-automatic or automatic statistics of the number of the ultrasonic fetal ribs and vertebrae.

After the identification module 1002 determines the position of the rib structure or spine of the fetus in the volume data, the statistical module 1003 identifies and marks the 12 ribs, spine and lumbar vertebrae of the fetus by a semi-automatic or automatic method, calculates the number of the ribs of the fetus, and displays the marking results of the ribs, spine and lumbar vertebrae on the display module 1005.

It should be noted that the identification, position and name of the lumbar vertebrae are marked to serve as a reference object for the position of the spinal cord cone end, and the position and ascending rule of the fetal spinal cord cone end are usually determined clinically by using the specific vertebral body level corresponding to the spinal cord cone end, for example, the normal adult spinal cord cone end is located at the level of lumbar vertebrae L1-2. The calculation and labeling of the number and the position of the lumbar vertebrae are not used as judgment indexes for congenital disease examination, but are only used as reference and measurement indexes for representing specific positions of the conical ends of the spinal cord of the fetus.

1. Counting the number of the rib bones of the semi-automatic fetus:

the following methods are adopted on the basis of the fetal rib structure body data segmented by the identification module 1002 in combination with information provided by a user through a certain manual operation workflow: the method comprises the following steps of counting the number of the ribs or the vertebral bodies of the fetus by one or more of a gray histogram projection method, a contour extraction algorithm, an edge extraction algorithm, a communication domain method, a blob detection algorithm, a template matching algorithm, an image feature extraction algorithm and a morphological operation algorithm, and marking the name of each rib or each vertebral body.

For example, first, 1 or more ribs or vertebral body structures are selected in segmented fetal rib or vertebral body data through a trackball, a touch screen or other tools, for example, one point on the T3 and T9 ribs or the central lines of the T3 and T9 ribs are respectively drawn, that is, the two ribs are marked as known, then, the results obtained after the segmentation of the costal body data and the fixed arrangement sequence relationship and spatial position relationship between the T1 and T12 ribs are used, for example, the rib above the T2 rib is always T3 below the T1 under the condition that the rib structures are not missing, and the above algorithm is combined to perform statistics and name marking.

For example, a certain coronal plane of the segmented volume data is projected along the Y-axis by the histogram gray projection method, the number and the position of the fetal ribs are respectively obtained by calculating the number of pixel peaks and the positions of the peak values of the pixel statistical map, the distance calculation is performed by the approximate area information of the drawn points and the drawn lines and the obtained position information, the closest ribs are marked as T3 and T9, and all other ribs are sequentially marked with names or only the ribs between T3 and T9.

Similarly, the position of each segmented rib can be determined according to the contour searching mode, and the number of the outermost surrounding contour, namely the maximum contour, to the ribs is calculated, so that some image preprocessing work may be required to remove some irrelevant structures to reduce interference, and the function can be realized by calculating the number and the position of the specific contour. Blob detection, template matching, and the like, and will not be described in detail.

2. Method for automatically counting fetal ribs or vertebral bodies

After segmenting the volume data of the fetal rib structure or spine, the number of fetal ribs in the volume data can be directly calculated by using the known spatial position relationship between the fetal ribs, such as the arrangement sequence between the ribs (T1-T12) shown in fig. 11, or some other known a priori knowledge and combining with the image processing method including, but not limited to, one or more methods such as image edge extraction, histogram image gray projection statistics, image contour extraction, morphology processing, threshold segmentation, blob detection, and the like. Here, in the arrangement order of the ribs shown in fig. 11, only T3, T6, T9, and T12 are labeled, that is, only the 3 rd rib, the 6 th rib, the 9 th rib, and the 12 th rib are labeled.

For example, the extracted fetal rib structure or coronal section image of the spine segmented by the machine learning or segmentation algorithm in the identification module 1002 is first subjected to threshold segmentation, then histogram statistical projection of gray values is performed along the direction perpendicular to the spine or parallel to the direction of the spine, the vertical coordinate is the image height, the horizontal coordinate is the number of pixels, then the number and position of the fetal ribs can be obtained by setting the number of peaks of the threshold calculation statistical map, and then marking can be performed, and similarly, the position of the spine can also be obtained for marking.

As another example, the number of ribs and the like can be directly obtained by combining contour extraction and blob detection.

A machine learning method can also be adopted, 12 different ribs and spines are taken as an identification target to construct a learning database, the learning database can distinguish the characteristics or rules of each specific rib and other ribs or spines, the characteristics or rules of each different rib and spine are learned to position and identify the ribs and spines in the fetal rib structure in other body data, wherein a calibration result can be set according to the actual task requirement, can be an ROI box containing the target, and can also be a Mask (Mask) for accurately segmenting the target; here, each rib or spine of the fetus is taken as a different class to identify and locate the target, and the class of the rib or vertebral body of each ROI box or Mask needs to be specified to identify and locate the problem for multiple targets. After the database is constructed, a machine learning algorithm is designed to learn the characteristics or rules of the fetal rib region and the non-fetal rib region in the fetal rib database so as to realize the positioning and identification of the fetal ribs in the volume data.

The general method is similar to the identification module 1002, such as: the first method is a traditional method for carrying out feature extraction and classification by a discriminator based on a sliding window. And a second method is a Bounding-Box method detection and identification method based on deep learning. And thirdly, an end-to-end semantic segmentation network method based on deep learning. And the fourth method is a method for positioning the target by only adopting the first method, the second method or the third method and then classifying and judging the target of the positioning result by the classifier.

Different ribs and spines can also be detected in the fetal rib structure data by adopting a template matching method, for example, different templates are established according to each different rib and spine of a fetus, all possible regions of the extracted fetal rib structure are traversed and segmented during detection, similarity matching is carried out on the template, the region with the highest similarity is selected as a target region, and then marking and quantity statistics are carried out according to matching results.

By a combination of methods including, but not limited to, one or more of the above, rib structures in the identified fetal rib structures can be counted and the number of ribs marked and counted.

It should be noted that the statistical module 1003 performs statistics on the number of vertebral bodies in the spine and the number of fetal ribs in the fetal rib structure, so as to perform statistics on the number of vertebral bodies included in the spine and the identifier of each vertebral body, thereby determining the lumbar vertebrae.

An imaging module 100 for automated imaging of fetal rib structures.

When the fetal rib structure is automatically imaged, the straightening algorithm is used for straightening and reconstructing the data of the rib structure body, and then a required volume data section is identified and positioned for imaging. The main contents of automatic imaging as shown in fig. 11 are automatic imaging of coronal plane of ribs or straightened ribs, automatic imaging of transverse plane of each rib or straightened ribs, automatic imaging of transverse plane of manually selected ribs, automatic imaging of three-dimensional rib cage extraction, and automatic imaging of vertebra VR chart labeled with spinal cord conical end position.

The volume data of the ribs and the volume data of the spine need to be straightened and reconstructed to obtain complete coronal and transverse planes. Straightening of the straightening object comprises two parts of extraction of a longitudinal axis and straightening reconstruction, wherein the straightening object comprises ribs and a spine.

The method of extracting the longitudinal axis of the straightened object may be: a longitudinal axis extraction algorithm based on tracking, a multi-scale longitudinal axis extraction algorithm based on a model, a longitudinal axis extraction method based on morphology, a centerline extraction method based on region growing, a method based on three-dimensional geometrical moments, a method for locating a centerline by machine learning, and the like.

The longitudinal axis extraction algorithm based on tracking is a semi-automatic algorithm, a tangent plane perpendicular to the tracking direction is generated in the tracking process based on initial key points and end points provided by user interaction, the central point of a rib or a spine in the tangent plane is accurately calculated by adopting a maximum likelihood value method and a centroid method, and after the tracking is finished, interpolation fitting is carried out on a central point sequence to obtain the longitudinal axis. At the moment, the longitudinal axis is sampled at equal intervals to generate an equal-interval rib or spine section sequence perpendicular to the rib or spine direction, and finally, the equal-interval section is reconstructed to achieve the purpose of straightening the rib or spine.

The method comprises the steps of approximating a local rib or spine to a tubular structure by a multi-scale longitudinal axis extraction algorithm based on a model, taking the gravity center of the tubular structure obtained by calculating geometric moments as the center of the local rib or spine, analyzing the eigenvalue of a Hessian matrix corresponding to a certain voxel under multi-scale Gaussian filtering, enhancing the local rib or vertebra structure, and estimating the longitudinal axis direction according to the eigenvector corresponding to the minimum eigenvalue of the Hessian matrix. After the longitudinal axis is obtained, the rib and the vertebra can be straightened and reconstructed, the longitudinal axis of the rib is sampled at equal intervals to obtain an equal-interval central point, and then a section sequence which is at equal intervals and vertical to the direction of the rib or the vertebra is regenerated on the basis, and the section sequence are stacked together to obtain rib section pictures or three-dimensional reconstruction pictures at different angles, so that the purpose of straightening is achieved.

The imaging of the coronal plane of the rib, the imaging of the cross-section of the rib, the imaging of the rib cage and the labeling of the spinal cone will be described below. The object imaged in the imaging of the coronal plane of the rib and the imaging of the cross-section of the rib may be the rib or the straightened rib, among others.

Imaging of coronal plane of first, ribs

The coronal plane of the rib needs to be acquired only after the spine and the rib are straightened simultaneously, as shown in fig. 12, the rib and the spine need to be straightened along the long axis direction, as shown by an arrow 1201 and an arrow 1202 in the figure, after the data of the rib body is straightened towards the two directions shown by the arrows by adopting the straightening method, the longitudinal axes of all the extracted ribs and spine almost all lie on the same plane, the plane is determined directly by solving a mathematical equation or a plane fitting method like a least square method or Hough transformation, and imaging is performed according to the volume data of the plane, so that the coronal plane of the rib structure of the fetus is acquired.

Second, imaging of the Cross section of Ribs

The rib cross-section is shown by straightening the ribs along the long axis 1202 in fig. 12, and in order to accurately find a cross-section that defines all of the ribs, the spine is straightened in the direction of arrow 1202 in fig. 12. The transverse section of each rib is determined by a method of determining a plane by two straight lines. Wherein the transverse plane must be perpendicular to the longitudinal axis of the spine and coplanar with the longitudinal axis of the ribs, so that the transverse plane without further ribs can be uniquely determined.

The rib cross section display is divided into two display modes, namely, all rib cross section display shown in fig. 13 and automatic display of a manually selected specified rib cross section shown in fig. 14.

Third, rib cage imaging

The structure of each rib T1-T12 and the centerline of the spine are identified by the identification module 1002 or the statistics module 1003 and the identified ribs are marked.

First, the volume data identifying the rib and spine structures segmented is reconstructed into new volume data, such as: the gray scale of the fetal rib structure is 1, and the gray scale of the non-fetal rib structure is 0. Then, the reconstructed volume data is rendered by using a volume rendering or surface rendering method, so as to obtain a three-dimensional skeleton rendering map of the fetal rib and vertebra structure shown in fig. 15.

In practical applications, the cross section of each straightened rib and the three-dimensional skeleton can be displayed on the same display interface.

Fourth, marking of the spinal cord Cone

The specific position (two-dimensional coordinate) of the spinal cord cone end located by the identification module 02 on the sagittal plane is combined with the specific position of the sagittal plane in the volume data to calculate a three-dimensional geometric coordinate point of the spinal cord cone end in the volume data, and finally a plane or a straight line which passes through the spinal cord cone end point and is perpendicular to the spinal column is calculated according to the three-dimensional coordinate point and the posture (such as the obtained central line of the spinal column) of the fetal spinal column located by the identification module in the volume data. The plane or the straight line is mapped to a three-dimensional VR diagram of the vertebral body structure, so that the relative position of the conical end of the spinal cord relative to the lumbar vertebrae of the fetus can be visually shown, and the result (such as the distance between L1 and L2) is calculated and marked on the VR diagram.

In practical application, the detection of fetal ribs, lumbar vertebrae and spinal cord cones can be carried out simultaneously, and the ribs and the spinal cord cones are marked on VR images comprising fetal rib structures and vertebral body structures. As shown in fig. 16, the vertebral positions corresponding to the particular location of the fetal spinal cord cone located are indicated by line 1701.

The application provides an ultrasonic detection method for abnormal number of ribs and abnormal position of spinal cord cone of a fetus. After obtaining the three-dimensional volume data of the fetus, automatically or semi-automatically (manually appointing 1 or 2 ribs) identifying, positioning and segmenting the structure of the ribs of the fetus, automatically/semi-automatically positioning the accurate position of the spinal coniform of the fetus, carrying out automatic/semi-automatic statistics on the number of the ribs of the fetus, carrying out (three-dimensional) extraction on a rib framework from the volume data, automatically imaging a cross section, a coronal plane and a stereo VR (virtual reality) diagram of all the ribs or appointed ribs after being straightened, and marking the specific position of the spinal coniform on the lumbar vertebra on the stereo VR diagram.

The method greatly simplifies the workflow of fetal rib examination and spinal cord conical position examination, can liberate doctors from complicated manual operation, puts more energy into disease diagnosis, reduces the dependence on the technology of the doctors, and improves the examination efficiency; and compared with the stability and imaging quality of the rib number statistical result and the spinal cone positioning result, the stability and imaging quality of the rib number statistical result and the spinal cone positioning result are in a better state manually, and misdiagnosis and missed diagnosis rate are reduced.

In one embodiment, an ultrasound imaging method is also provided. The ultrasonic imaging method can be applied to the ultrasonic imaging apparatus described previously. In the method, three-dimensional volume data of the ribs of the fetus can be automatically or semi-automatically identified from the three-dimensional volume data of the fetus based on the characteristics of the ribs of the fetus, and clinically valuable sectional images, such as coronal images, cross-sectional images, longitudinal sectional images of the ribs, and the like, are then automatically or semi-automatically obtained based on the three-dimensional volume data of the identified ribs of the fetus.

For example, in this embodiment, three-dimensional volume data of the fetus may be acquired first. The three-dimensional volume data of the fetus may be obtained by real-time scanning by an ultrasound imaging device, for example, the ultrasound imaging device transmits an ultrasonic wave to the fetus through an ultrasound probe and receives an ultrasonic echo to obtain an ultrasonic echo signal, and a processor of the ultrasound device processes the ultrasonic echo signal to obtain the three-dimensional volume data of the fetus. The three-dimensional volume data of the fetus can also be acquired and stored in advance by the ultrasonic imaging device, and read in for processing when a section image or a three-dimensional image of the rib of the fetus needs to be acquired.

The processor may then identify three-dimensional volumetric data of the ribs of the fetus from the three-dimensional volumetric data of the fetus based on the characteristics of the ribs of the fetus. Here, the fetal rib may show a specific feature on the ultrasound image due to its own characteristics. Thus, the processor may identify three-dimensional volumetric data of the fetal rib, such as a mean, variance, distribution features, texture, morphological features, etc., of gray scale or pixel or voxel values based on the image features of the fetal rib (i.e., the features exhibited by its image data).

Then, the processor may obtain a first plane (e.g., coronal plane) or a first curved plane passing through at least two of the three-dimensional volume data of the fetal ribs and being parallel to or coincident with an arrangement plane of the plurality of fetal ribs in the three-dimensional volume data of the fetal ribs and/or obtain a second plane (e.g., cross section) or a second curved plane passing through at least one of the three-dimensional volume data of the fetal ribs and intersecting with the arrangement plane of the plurality of fetal ribs in the three-dimensional volume data of the fetal ribs from the three-dimensional volume data of the identified fetal ribs, and then obtain an image on the first plane or the first curved plane and/or obtain an image on the second plane or the second curved plane from the three-dimensional volume data of the identified fetal ribs.

After obtaining the image on the first plane or the first curved surface and/or the image on the second plane or the second curved surface, the processor may display the image on the first plane or the first curved surface as a two-dimensional image and/or the image on the second plane or the second curved surface as a two-dimensional image through the display.

In this embodiment, the method may further include: and obtaining a three-dimensional ultrasonic image of the fetal rib according to the identified three-dimensional volume data of the fetal rib, and displaying the three-dimensional ultrasonic image of the fetal rib.

In this embodiment, the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs may be automatically identified from the three-dimensional volume data of the first tissue to be detected based on the first rib detection model.

In this embodiment, the template matching method may also be used to identify the three-dimensional volume data of the fetal rib from the three-dimensional volume data of the fetus based on the features of the fetal rib. For example, at least two first candidate regions may be determined from the three-dimensional volume data of the fetus, the volume data feature of the three-dimensional volume data of each first candidate region is obtained, the first matching degree between each first candidate region and the ribs of the fetus is determined according to the volume data feature of each first candidate region, the first candidate region with the highest first matching degree is determined to be the target region corresponding to the ribs of the fetus, and the three-dimensional volume data of the target region corresponding to the ribs of the fetus is used as the three-dimensional volume data of the ribs of the fetus.

In this embodiment, when identifying the three-dimensional volume data of the fetal rib, the identification may be performed based on an input operation of a user or another device connected to the ultrasound imaging device through a wired or wireless network, that is, may be performed semi-automatically. For example, a three-dimensional ultrasound image corresponding to the three-dimensional volume data of the fetus may be displayed, a first input operation may be received based on the three-dimensional ultrasound image corresponding to the three-dimensional volume data of the fetus, a landmark point corresponding to the first input operation may be determined, and the three-dimensional volume data of the rib of the fetus may be identified from the three-dimensional volume data of the fetus according to the coordinates of the landmark point.

In this embodiment, the overall structures of the fetal ribs and the fetal spine may be identified first, and then the fetal ribs may be identified. For example, three-dimensional volume data of a fetal rib structure including a fetal rib and a fetal spine may be identified from three-dimensional volume data of a fetus, and then three-dimensional volume data of a fetal rib is identified from the three-dimensional volume data of the fetal rib structure.

In this embodiment, the ribs may be straightened, and then an image on the first plane or the first curved surface and/or an image on the second plane may be obtained based on the three-dimensional volume data of the straightened ribs. For example, the three-dimensional volume data of the identified fetal ribs may be straightened to obtain straightened rib three-dimensional volume data, and then a first plane or a first curved surface which passes through at least two ribs in the straightened rib three-dimensional volume data and is parallel to or coincident with the arrangement surface of the plurality of fetal ribs in the straightened rib three-dimensional volume data is obtained according to the straightened rib three-dimensional volume data, and/or a second plane which passes through at least one rib in the straightened rib three-dimensional volume data and intersects with the arrangement surface of the plurality of fetal ribs in the straightened rib three-dimensional volume data is obtained, and an image on the first plane or the first curved surface and/or an image on the second plane is obtained according to the straightened rib three-dimensional volume data.

In this embodiment, the fetal spine may also be identified, and after straightening both the fetal ribs and the fetal spine, images on the first plane and/or the second plane may be obtained from the straightened three-dimensional volume data. For example, the three-dimensional volume data of the fetal spine may be identified from the three-dimensional volume data of the fetus based on the characteristics of the fetal spine, and the identified fetal ribs and the identified fetal spine are generally connected to each other, so the fetal ribs and the fetal spine are collectively referred to as a fetal rib structure herein. In this embodiment, the three-dimensional volume data of the fetal rib structure may be straightened, that is, both the three-dimensional volume data of the fetal rib and the three-dimensional volume data of the fetal spine are straightened, to obtain the straightened rib structure three-dimensional volume data, and accordingly, the straightened rib structure three-dimensional volume data includes the straightened rib three-dimensional volume data and the straightened spine three-dimensional volume data. Then, according to the three-dimensional volume data of the straightened rib structure, a first plane passing through the straightened ribs and the straightened spine and/or a second plane passing through at least one straightened rib and intersecting with the straightened spine are obtained, and according to the three-dimensional volume data of the straightened rib structure, an image on the first plane is obtained, and/or according to the three-dimensional volume data of the straightened rib structure, an image on the second plane is obtained.

In this embodiment, when the image on the first plane is obtained, the image on the first plane may be obtained based on data within a certain thickness range, so that the image on the first plane may be enabled to display more information. For example, three-dimensional volume data within a predetermined thickness range in a direction perpendicular to the first plane may be acquired from the straightened rib three-dimensional volume data and/or the straightened spine three-dimensional volume data, and an image on the first plane may be obtained from the three-dimensional volume data within the predetermined thickness range.

The three-dimensional volume data within the predetermined thickness range may be manually determined by a user. For example, the user may directly draw a predetermined thickness range on the displayed three-dimensional image or two-dimensional sectional image of the straightened three-dimensional volume data through the input device, or may determine a predetermined plane (for example, a plane where the upper and lower edges of the vertebral arch or vertebral body are located, or other planes according to actual needs, etc.) on the displayed three-dimensional image or two-dimensional sectional image of the straightened three-dimensional volume data through the input device, and then the region between the upper and lower edge planes is the predetermined thickness range, etc.

In this embodiment, the three-dimensional volumetric data within the predetermined thickness range may be made to include the vertebral arch and/or vertebral body of the fetal spine. At this time, too, the predetermined thickness range may be manually set so that it includes the vertebral arch and/or vertebral bodies of the fetal spine. Alternatively, the predetermined thickness range may be determined automatically or semi-automatically. For example, the vertebral arch and/or the vertebral body can be identified from the straightened spine three-dimensional volume data based on the characteristics of the vertebral arch and/or the vertebral body of the spine, and the three-dimensional volume data in the preset thickness range can be determined so that the three-dimensional volume data in the preset thickness range contains the identified vertebral arch and/or the vertebral body, for example, three planes of upper and lower edges can be fitted according to the upper and lower edges of the identified vertebral arch and/or the vertebral body, and the range between any two planes in the three planes is the preset thickness range.

In this embodiment, the method for identifying the vertebral arch and/or the vertebral body from the straightened spine three-dimensional volume data based on the characteristics of the vertebral arch and/or the vertebral body of the spine may refer to the method for identifying the specific tissue structure from the three-dimensional volume data in the foregoing embodiments, and similar methods may be used, and will not be described in detail herein.

In this embodiment, the image on the first plane may be obtained from three-dimensional volume data within a predetermined thickness range using a variety of suitable means. For example, three-dimensional volume data within a predetermined thickness range may be weighted in a direction perpendicular to a first plane, an image on the first plane is obtained, and so on. For example, weighted average may be performed on all voxels in a path perpendicular to the first plane in the thickness range, to obtain values of pixel points corresponding to the path in the first plane in the direction, and similarly obtain values of all pixel points in the first plane, that is, to obtain an image in the first plane.

In this embodiment, the three-dimensional volume data of the identified fetal rib can be straightened using a variety of suitable methods to obtain straightened rib three-dimensional volume data. For example, a longitudinal axis of the three-dimensional volume data of the identified fetal rib can be determined, the three-dimensional volume data of the identified fetal rib can be sampled according to the longitudinal axis to obtain a section sequence, and then the section sequence is reconstructed along a straight line to obtain the three-dimensional volume data of the straightened rib.

In this embodiment, similarly, the three-dimensional volume data of the identified fetal spine may be straightened using a variety of suitable methods to obtain straightened spine three-dimensional volume data. For example, a longitudinal axis of the identified three-dimensional volume data of the fetal spine may be determined, the identified three-dimensional volume data of the fetal spine may be sampled according to the longitudinal axis to obtain a cut sequence, and the cut sequence may be reconstructed along a straight line to obtain straightened spine three-dimensional volume data.

In this embodiment, the number of the fetal ribs may also be automatically determined according to the three-dimensional volume data of the identified fetal ribs, and displayed. The number of fetal ribs may be displayed as a number or other suitable symbol.

In this embodiment, the fetal rib may also be marked according to the three-dimensional volume data of the identified fetal rib, and the mark of the fetal rib is displayed. The indicia may be any suitable indicia, such as the aforementioned T1, T2, etc. indicia, or any other suitable text, numbers, symbols, colors, etc.

In this embodiment, specific schemes of the steps involved may refer to the methods in the foregoing embodiments or be the same as or similar to the similar steps in the foregoing embodiments, and detailed descriptions thereof are omitted here.

In one embodiment, an ultrasound imaging method is provided, which may be applied to the aforementioned ultrasound imaging apparatus. The method can comprise the following steps: acquiring three-dimensional volume data of a fetus, identifying a spinal cord conical region from the three-dimensional volume data of the fetus based on the characteristics of the spinal cord conical region of the fetus, determining the position of the spinal cord conical region according to the identified spinal cord conical region, and displaying the position of the spinal cord conical region. Here, the location of the spinal conical region may be displayed in various suitable ways, such as by suitable symbols, colored regions, text, arrows, geometric shapes, and so forth.

In this embodiment, the location of the identified conical region of the spinal cord may be the location of the conical end of the spinal cord. For example, the location of the conical end of the spinal cord can be determined from the identified conical region of the spinal cord and displayed. The location of the conical end of the spinal cord may be displayed in a variety of suitable ways, such as, for example, a suitable composition, color, dots, lines, arrows, numbers, distance from a suitable reference location, and the like.

In this embodiment, the spinal conical region can be identified by a target matching method. For example, at least two second candidate regions may be determined from the three-dimensional volume data of the fetus, the volume data feature of the three-dimensional volume data of each second candidate region may be obtained, the second matching degree between each second candidate region and the spinal conus may be determined according to the volume data feature of each second candidate region, and the second candidate region with the highest second matching degree may be determined as the spinal conus region.

In this embodiment, the sagittal plane of the fetus may be identified from the three-dimensional volume data of the fetus, and then the spinal conical area may be identified from the sagittal plane image. For example, a sagittal plane image of the spinal column passing through the fetus can be determined from the three-dimensional volumetric data of the fetus based on the features of the sagittal plane of the spinal column passing through the fetus, and then the spinal cone region can be determined in the sagittal plane image of the spinal column passing through the fetus based on the features of the spinal cone. Here, the sagittal plane may be the median sagittal plane and/or a sagittal plane adjacent to the median sagittal plane.

In this embodiment, the lumbar region can be identified from the three-dimensional volume data of the fetus, and the position of the spinal cone region is displayed relative to the lumbar region, so that the user can easily see the relative positional relationship between the spinal cone and the lumbar. For example, a lumbar region may be identified from the three-dimensional volumetric data of the fetus based on features of the lumbar spine of the fetus, an ultrasound image of the lumbar region is displayed, and the location of a spinal conical region (e.g., the end of the spinal cone, etc.) is displayed relative to the ultrasound image of the lumbar region. Here, displaying the position of the spinal conic region with respect to the ultrasound image of the lumbar region may include various suitable manners, for example, the ultrasound image of the lumbar region and the position of the spinal conic region may be displayed simultaneously so that the user may directly see the relative positional relationship therebetween, or the distance of the spinal conic region with respect to the lumbar region may be displayed by characters or symbols or the like, or the relative positional relationship between the two may be displayed by symbols representing the lumbar and spinal conic regions, or the like.

In this embodiment, the specific schemes of the steps involved (e.g., identifying the spinal conical region, identifying the lumbar region, identifying the sagittal plane of the fetus, etc.) can be the same as or similar to the methods in the previous embodiments, and are not described in detail herein.

An embodiment of the present application further provides an ultrasound imaging apparatus 10, as shown in fig. 1, including:

a probe head 100;

a transmitting circuit 101 for exciting the probe 100 to transmit ultrasonic waves to a first tissue to be detected;

a receiving circuit 103 for receiving an ultrasonic echo returned from the first tissue to be tested through the probe 100 to obtain an ultrasonic echo signal;

a processor 105, which processes the ultrasound echo signals to obtain a three-dimensional ultrasound image of the marked fetal rib structure;

a display 106 for displaying the marked three-dimensional ultrasonic image of the fetal rib structure;

wherein, the processor 105 further executes the following steps:

acquiring three-dimensional volume data of a first tissue to be detected according to the ultrasonic echo information; identifying three-dimensional volume data of a vertebral column of a fetus and three-dimensional volume data of ribs of the fetus from the three-dimensional volume data of the first tissue to be detected; rendering the three-dimensional volume data of the fetal spine and the three-dimensional volume data of the fetal ribs to obtain a three-dimensional ultrasonic image of the fetal rib structure; marking the fetal ribs in the three-dimensional ultrasound image of the fetal rib structure; and outputting the marked three-dimensional ultrasonic image of the fetal rib structure.

a probe head 100;

the transmitting circuit 101 is used for exciting the probe 100 to transmit ultrasonic waves to a second tissue to be detected;

a receiving circuit 103, configured to receive an ultrasonic echo returned from the second tissue to be tested through the probe 100 to obtain an ultrasonic echo signal;

a processor 105 for processing the ultrasound echo signals to obtain a three-dimensional ultrasound image of the marked vertebral structure;

a display 106 for displaying a three-dimensional ultrasound image of the marked vertebral structure;

wherein the processor 105 further performs the following steps:

acquiring three-dimensional volume data of a second tissue to be detected according to the ultrasonic echo information; identifying three-dimensional volume data of spinal cord cones and three-dimensional volume data of lumbar vertebrae from the second three-dimensional volume data; rendering the three-dimensional data of the spinal cone and the three-dimensional data of the lumbar vertebra to obtain a three-dimensional ultrasonic image of the vertebra structure; marking the spinal cone in a three-dimensional ultrasound image of the vertebral structure; outputting a three-dimensional ultrasound image of the marked vertebral structure.

Accordingly, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the steps of the ultrasound imaging method described above.

The above description of the ultrasound imaging system and computer readable storage medium embodiments is similar to the description of the method embodiments above with similar beneficial results as the method embodiments. For technical details not disclosed in embodiments of the ultrasound imaging system and computer readable storage medium of the present invention, reference is made to the description of embodiments of the method of the present invention for understanding.

In the embodiment of the present application, if the ultrasound imaging method is implemented in the form of a software functional module and sold or used as a standalone product, the ultrasound imaging method may also be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application may be essentially implemented or portions thereof contributing to the prior art may be embodied in the form of a software product stored in a storage medium, and including several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read Only Memory (ROM), a magnetic disk, or an optical disk, and various media capable of storing program codes. Thus, embodiments of the present application are not limited to any specific combination of hardware and software.

It should be appreciated that reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present application. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. It should be understood that, in the various embodiments of the present application, the sequence numbers of the above-mentioned processes do not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application. The above-mentioned serial numbers of the embodiments of the present application are merely for description, and do not represent the advantages and disadvantages of the embodiments.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or apparatus that comprises the element.

In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of the unit is only a logical functional division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or other forms.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units; can be located in one place or distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, all functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.

Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium includes: various media that can store program codes, such as a removable Memory device, a Read Only Memory (ROM), a magnetic disk, or an optical disk.

Alternatively, the integrated units described above in the present application may be stored in a computer-readable storage medium if they are implemented in the form of software functional modules and sold or used as independent products. Based on such understanding, the technical solutions of the embodiments of the present application may be essentially implemented or portions thereof contributing to the prior art may be embodied in the form of a software product stored in a storage medium, and including several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a removable storage device, a ROM, a magnetic or optical disk, or other various media that can store program code.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. An ultrasonic imaging method is applied to an ultrasonic imaging device, and the method comprises the following steps:

acquiring three-dimensional volume data of a fetus;

identifying three-dimensional volume data of the ribs of the fetus from the three-dimensional volume data of the fetus based on the characteristics of the ribs of the fetus;

according to the identified three-dimensional volume data of the fetal ribs, obtaining a first plane or a first curved surface which passes through at least two ribs in the three-dimensional volume data of the fetal ribs and is parallel to or coincided with the arrangement surface of a plurality of fetal ribs in the three-dimensional volume data of the fetal ribs and/or obtaining a second plane or a second curved surface which passes through at least one rib in the three-dimensional volume data of the fetal ribs and is intersected with the arrangement surface of the plurality of fetal ribs in the three-dimensional volume data of the fetal ribs;

obtaining an image on the first plane or the first curved surface and/or obtaining an image on the second plane or the second curved surface according to the identified three-dimensional volume data of the ribs of the fetus;

and displaying the image on the first plane or the first curved surface as a two-dimensional image and/or displaying the image on the second plane or the second curved surface as a two-dimensional image.

2. The method of claim 1, further comprising:

obtaining a three-dimensional ultrasonic image of the fetal rib according to the identified three-dimensional volume data of the fetal rib;

and displaying the three-dimensional ultrasonic image of the fetal rib.

3. The method of claim 1 or 2, wherein identifying three-dimensional volume data of the ribs of the fetus from the three-dimensional volume data of the fetus based on the features of the ribs of the fetus comprises:

and identifying the three-dimensional volume data of the spine of the fetus and the three-dimensional volume data of the ribs of the fetus from the three-dimensional volume data of the first tissue to be detected based on a first rib detection model.

4. The method of claim 1 or 2, wherein identifying three-dimensional volume data of the ribs of the fetus from the three-dimensional volume data of the fetus based on the features of the ribs of the fetus comprises:

determining at least two first candidate regions from the three-dimensional volume data of the fetus, and acquiring the volume data characteristics of the three-dimensional volume data of each first candidate region;

determining a first matching degree of each first candidate region and the ribs of the fetus according to the volume data characteristics of each first candidate region;

determining a first candidate region with the highest first matching degree as a target region corresponding to the ribs of the fetus;

and taking the three-dimensional volume data of the target area corresponding to the fetal rib as the three-dimensional volume data of the fetal rib.

5. The method of claim 1 or 2, wherein identifying the three-dimensional volume data of the ribs of the fetus from the three-dimensional volume data of the fetus based on the features of the ribs of the fetus comprises:

displaying a three-dimensional ultrasonic image corresponding to the three-dimensional volume data of the fetus;

receiving a first input operation based on a three-dimensional ultrasonic image corresponding to the three-dimensional volume data of the fetus;

determining a mark point corresponding to the first input operation;

and identifying the three-dimensional volume data of the ribs of the fetus from the three-dimensional volume data of the fetus according to the coordinates of the marking points.

6. The method of claim 1 or 2, wherein identifying the three-dimensional volume data of the ribs of the fetus from the three-dimensional volume data of the fetus based on the features of the ribs of the fetus comprises:

identifying three-dimensional volumetric data of a fetal rib structure from the three-dimensional volumetric data of the fetus, wherein the fetal rib structure comprises a fetal rib and a fetal spine;

three-dimensional volume data of the ribs of the fetus is identified from the three-dimensional volume data of the structures of the ribs of the fetus.

7. The method according to any one of claims 1 to 6,

according to the identified three-dimensional volume data of the fetal ribs, obtaining a first plane or a first curved surface which passes through at least two of the three-dimensional volume data of the fetal ribs and is parallel to or coincident with an arrangement plane of a plurality of fetal ribs in the three-dimensional volume data of the fetal ribs and/or obtaining a second plane or a second curved surface which passes through at least one of the three-dimensional volume data of the fetal ribs and intersects with the arrangement plane of the plurality of fetal ribs in the three-dimensional volume data of the fetal ribs, the method comprises the following steps:

straightening the three-dimensional data of the identified ribs of the fetus to obtain straightened rib three-dimensional data;

according to the straightening rib three-dimensional volume data, obtaining a first plane or a first curved surface which passes through at least two ribs in the straightening rib three-dimensional volume data and is parallel to or coincident with an arrangement plane of a plurality of fetal ribs in the straightening rib three-dimensional volume data, and/or obtaining a second plane which passes through at least one rib in the straightening rib three-dimensional volume data and is intersected with the arrangement plane of the plurality of fetal ribs in the straightening rib three-dimensional volume data;

obtaining an image on the first plane or first curved surface and/or obtaining an image on the second plane or second curved surface from the identified three-dimensional volumetric data of the fetal ribs, including

And obtaining an image on the first plane or the first curved surface and/or obtaining an image on the second plane according to the straightened rib three-dimensional volume data.

8. The method according to any one of claims 1 to 6,

according to the identified three-dimensional volume data of the ribs of the fetus, obtaining a first plane or a first curved surface which passes through at least two ribs in the three-dimensional volume data of the ribs of the fetus and is parallel to or coincident with an arrangement plane of a plurality of ribs of the three-dimensional volume data of the ribs of the fetus and/or obtaining a second plane or a second curved surface which passes through at least one rib in the three-dimensional volume data of the ribs of the fetus and intersects with the arrangement plane of the plurality of ribs of the fetus in the three-dimensional volume data of the ribs of the fetus comprises the following steps:

identifying three-dimensional volume data of a spine of a fetus from the three-dimensional volume data of the fetus based on features of the spine of the fetus;

straightening three-dimensional volume data of a fetal rib structure to obtain straightened rib structure three-dimensional volume data, wherein the three-dimensional volume data of the fetal rib structure comprise the three-dimensional volume data of the fetal ribs and the three-dimensional volume data of the fetal spine, and the straightened rib structure three-dimensional volume data comprise straightened rib three-dimensional volume data and straightened spine three-dimensional volume data;

obtaining a first plane passing through the straightened ribs and the straightened spine and/or obtaining a second plane passing through at least one straightened rib and intersecting with the straightened spine according to the three-dimensional volume data of the straightened rib structure; obtaining an image on the first plane or first curved surface and/or obtaining an image on the second plane or second curved surface from the identified three-dimensional volumetric data of the fetal ribs, including

And obtaining an image on the first plane according to the straightened rib structure three-dimensional volume data, and/or obtaining an image on the second plane according to the straightened rib structure three-dimensional volume data.

9. The method of claim 8, wherein obtaining the image on the first plane from the straightened rib three-dimensional volume data and/or the straightened spine three-dimensional volume data comprises:

acquiring three-dimensional volume data within a preset thickness range in the direction perpendicular to the first plane from the straightened rib three-dimensional volume data and/or the straightened spine three-dimensional volume data;

and obtaining an image on the first plane according to the three-dimensional volume data in the preset thickness range.

10. The method of claim 8, wherein obtaining the image on the first plane from the three-dimensional volume data within the predetermined thickness range comprises: and weighting the three-dimensional volume data in the preset thickness range in the direction vertical to the first plane to obtain an image on the first plane.

11. The method of claim 9 or 10, further comprising:

identifying vertebral arches and/or vertebral bodies from the straightened spine three-dimensional volume data based on characteristics of the vertebral arches and/or vertebral bodies of the spine;

determining three-dimensional volumetric data within the predetermined thickness range such that the three-dimensional volumetric data within the predetermined thickness range includes the identified vertebral arch and/or vertebral body.

12. The method of any one of claims 7 to 11, wherein straightening the three-dimensional volume data of the identified fetal rib to obtain straightened rib three-dimensional volume data comprises:

determining a longitudinal axis of the three-dimensional volume data of the identified fetal ribs;

sampling the three-dimensional data of the identified fetal ribs according to the longitudinal axis to obtain a section sequence;

and reconstructing the tangent plane sequence along a straight line to obtain the three-dimensional volume data of the straightened rib.

13. The method of any one of claims 8 to 11, wherein straightening the identified three-dimensional volumetric data of the fetal spine to obtain straightened spinal three-dimensional volumetric data comprises:

determining a longitudinal axis of the identified three-dimensional volumetric data of the fetal spine;

sampling the three-dimensional data of the identified fetal spine according to the longitudinal axis to obtain a section sequence;

and reconstructing the tangent plane sequence along a straight line to obtain the straightened spine three-dimensional volume data.

14. The method according to any one of claims 1 to 13, further comprising:

determining the number of the ribs of the fetus according to the three-dimensional volume data of the identified ribs of the fetus;

displaying the number of fetal ribs;

and/or

Marking the fetal ribs according to the three-dimensional volume data of the identified fetal ribs;

displaying the marker of the fetal rib.