CN114266728B - Method, device, equipment and medium for acquiring magnetic resonance image - Google Patents
Method, device, equipment and medium for acquiring magnetic resonance image Download PDFInfo
- Publication number
- CN114266728B CN114266728B CN202111428699.8A CN202111428699A CN114266728B CN 114266728 B CN114266728 B CN 114266728B CN 202111428699 A CN202111428699 A CN 202111428699A CN 114266728 B CN114266728 B CN 114266728B
- Authority
- CN
- China
- Prior art keywords
- sampling
- space data
- magnetic resonance
- data
- relative
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000005070 sampling Methods 0.000 claims abstract description 204
- 238000005096 rolling process Methods 0.000 claims abstract description 6
- 238000003062 neural network model Methods 0.000 claims description 28
- 230000009466 transformation Effects 0.000 claims description 21
- 238000013519 translation Methods 0.000 claims description 10
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 238000013501 data transformation Methods 0.000 claims description 6
- 230000001131 transforming effect Effects 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims description 2
- 238000012795 verification Methods 0.000 description 8
- 238000013528 artificial neural network Methods 0.000 description 7
- 238000011156 evaluation Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000003042 antagnostic effect Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000002595 magnetic resonance imaging Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000002601 radiography Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The application provides a method, a device, equipment and a medium for acquiring magnetic resonance images, wherein the method comprises the following steps: acquiring sampling positions and K space data of each sampling moment of a magnetic resonance scanning part of a detected person, wherein the sampling positions comprise coordinate data, pitch angle, yaw angle and rolling angle; determining the relative position of each sampling position relative to a standard position; data alignment is carried out on the corresponding K space data according to the relative positions, and K space data after alignment is obtained; combining all the aligned K space data to obtain combined K space data; and performing inverse Fourier transform on the combined K space data to obtain a magnetic resonance image of a space domain, wherein the stability of artifact removal is excellent.
Description
Technical Field
The present application relates to the field of magnetic resonance image acquisition, and in particular, to a method, apparatus, device, and medium for acquiring magnetic resonance images.
Background
The magnetic resonance image is obtained by combining all K space data after acquiring K space data acquired for multiple times and then carrying out inverse Fourier transform according to the combined data. But the imaging speed of the magnetic resonance image is slow, and the fluctuation of the patient in the scanning process finally causes artifacts to appear in the magnetic resonance image obtained by combining the K space data acquired at different moments, thereby affecting the confirmation of the medical staff on the illness state of the patient.
The common means for solving the artifacts in the magnetic resonance image is a padding method, and a protection pad is added under the part to be inspected of the patient to reduce the fluctuation of the patient. In order to solve the above problems, the related art uses an artifact removal neural network model to perform artifact removal on an obtained magnetic resonance image with artifacts, specifically, trains a neural network module by using a magnetic resonance image with artifacts and a normal magnetic resonance image to obtain a trained artifact removal neural network model, and in practical application, inputs the magnetic resonance image with artifacts into the artifact removal neural network model to output the normal magnetic resonance image. By adopting the means to remove the artifacts, when the artifact removing neural network model automatically removes the artifacts, the obtained magnetic resonance image with the sign easily causes the distortion of the non-artifact part due to the uncertainty of the size of the artifacts, and the stability of the output result is poor.
In carrying out the present application, the applicant has found that at least the following problems exist in this technology: the artifact removal neural network model is easy to cause image distortion when processing artifact images, and the output stability is poor. Therefore, how to provide a solution to the above technical problem is a problem that a person skilled in the art needs to solve at present.
Disclosure of Invention
The application provides a magnetic resonance image acquisition method, which is realized by the following technical scheme:
a magnetic resonance image acquisition method comprising:
acquiring sampling positions and K space data of each sampling moment of a magnetic resonance scanning part of a detected person, wherein the sampling positions comprise coordinate data, pitch angle, yaw angle and rolling angle;
determining the relative position of each sampling position relative to a standard position;
Data alignment is carried out on the corresponding K space data according to the relative positions, and K space data after alignment is obtained;
combining all the aligned K space data to obtain combined K space data;
and performing inverse Fourier transform on the combined K space data to obtain a magnetic resonance image of a space domain.
According to the technical scheme, in the process of magnetic resonance scanning, each sampling moment is used for acquiring K space data and simultaneously acquiring the sampling position of the magnetic resonance scanning part, the magnetic resonance scanning part of a subject can fluctuate in the process of magnetic resonance scanning, and the sampling position of each sampling moment is different; after the standard position is determined, the relative position of each sampling position relative to the standard position is obtained, the relative position is used for aligning the K space data to the standard position, the movable K space data are restored, no motion artifact exists in the magnetic resonance image obtained after the K space data are finally combined, even the K space data with large motion amplitude can be restored, the image is not distorted, and the artifact removal stability is excellent.
Preferably, before determining the relative position of each sampling position with respect to the standard position, the method further includes:
And determining the standard position from all the sampling positions, or taking a set position as the standard position.
Through the technical scheme, the standard position is determined from all sampling positions, or a fixed sampling position is set as the standard position, so that the standard position can be flexibly set according to actual requirements.
Preferably, said determining said standard position from all said sampling positions comprises:
Determining the sampling position corresponding to the scanning of the middle sampling time in all the sampling times as the standard position;
Or determining the target sampling position with the same number of times and the maximum number of times from all the sampling positions, and determining the target sampling position as the standard position.
According to the technical scheme, the situation that data loss exists when K space data are subjected to space conversion occurs, and the sampling position corresponding to the middle sampling time or the target sampling position with the largest number of times determined in all the sampling positions is used as the standard position, so that movement can be reduced as much as possible, and error introduction is reduced.
Preferably, the data alignment is performed on the corresponding K-space data according to the relative position, so as to obtain aligned K-space data, including:
determining geometric transformation information according to the relative position, wherein the geometric transformation information comprises translation information and rotation information;
and carrying out data transformation on the K space data by utilizing Fourier transformation according to the geometric transformation information to obtain transformed K space data, and taking the transformed K space data as the aligned space data.
Through the technical scheme, after translation information and rotation information are determined, data transformation is carried out on K space data by utilizing Fourier transformation, data aligned with a standard position is obtained, and the data is used as aligned space data.
Preferably, the method further comprises:
the sampling frequency of the magnetic resonance scan is controlled to be one half of the interval of two pulses.
According to the technical scheme, when magnetic resonance scanning is carried out, tens of thousands of pulses are needed to acquire magnetic resonance images with enough definition, and the evaluation rate of sampling is close to half of the interval between two pulses during imaging, so that the accuracy of a simulation track is improved.
Preferably, the method further comprises:
And inputting the magnetic resonance image into a pre-established neural network model for high-quality conversion to obtain a high-quality magnetic resonance image.
Through the technical scheme, the neural network model for converting the low-quality image into the high-quality image is trained in advance, the magnetic resonance image is input into the neural network model, the high-quality magnetic resonance image is obtained, and the resolution of the image is improved.
Preferably, the acquiring the sampling position and K-space data of each sampling time of the magnetic resonance scanning part of the subject includes:
acquiring initial sampling positions and initial K space data of initial sampling moments of the magnetic resonance scanning part of the subject;
Acquiring a relative sampling position and corresponding current K space data of each sampling time of the magnetic resonance scanning part relative to the last sampling time;
And after all the relative sampling positions are acquired, transforming all the relative sampling positions by utilizing the initial sampling positions to obtain all the sampling positions, and obtaining all the K space data according to the initial K space data and all the current K space data.
Through the technical scheme, the relative sampling positions based on the last sampling moment are acquired each time, the data acquisition mode of the sampling moment is simplified, and then after sampling is completed, all the relative sampling positions are transformed by uniformly utilizing the initial sampling positions, so that all the sampling positions are obtained, and the final sampling position acquisition efficiency is improved.
The second object of the present application is to provide a magnetic resonance image acquisition apparatus, which is realized by the following technical scheme:
a magnetic resonance image acquisition device comprising:
The acquisition module is used for acquiring sampling positions and K space data of each sampling moment of a magnetic resonance scanning part of a detected person, wherein the sampling positions comprise coordinate data, pitch angle, yaw angle and rolling angle;
The relative position determining module is used for determining the relative position of each sampling position relative to the standard position;
The alignment module is used for carrying out data alignment on the corresponding K space data according to the relative position to obtain aligned K space data;
the merging module is used for merging all the K-space data after alignment to obtain merged K-space data;
And the Fourier transform module is used for carrying out inverse Fourier transform on the combined K space data to obtain a magnetic resonance image of a space domain.
The third object of the present application is to provide an electronic device, which is realized by the following technical scheme:
an electronic device comprising a memory and a processor, the memory having stored thereon a computer program capable of being loaded by the processor and performing a magnetic resonance image acquisition method as any one of the above.
The fourth object of the present application is to provide a computer storage medium, which is realized by the following technical scheme:
A computer readable storage medium having instructions stored therein, which when executed on a computer, cause the computer to perform a magnetic resonance image acquisition method as claimed in any one of the preceding claims.
Drawings
Fig. 1 is a flowchart of a magnetic resonance image acquisition method according to an embodiment of the present application;
Fig. 2 is a schematic structural diagram of a magnetic resonance image acquisition device according to an embodiment of the present application;
Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The present application will be described in further detail with reference to the accompanying drawings.
The present embodiment is only for explanation of the present application and is not to be construed as limiting the present application, and modifications which do not creatively contribute to the present embodiment may be made by those skilled in the art after reading the present specification as long as they are protected by patent laws within the scope of the embodiments of the present application.
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
In addition, the term "and/or" herein is merely an association relationship describing an association object, and means that three relationships may exist, for example, a and/or B may mean: a exists alone, A and B exist together, and B exists alone. In this context, unless otherwise specified, the term "/" generally indicates that the associated object is an "or" relationship.
The related art utilizes a neural network mode to train the neural network by using a magnetic resonance image with artifacts and a normal magnetic resonance image to obtain a trained artifact removal neural network model, and inputs the magnetic resonance image with the artifacts into the artifact removal neural network model to output the normal magnetic resonance image, but has poor stability and is easy to cause distortion of a non-artifact part.
In order to solve the technical problems, the embodiment of the application provides a magnetic resonance image acquisition method, wherein in the process of magnetic resonance scanning, each sampling moment acquires K space data and simultaneously acquires the sampling position of a magnetic resonance scanning part, the magnetic resonance scanning part of a detected subject can fluctuate in the process of magnetic resonance scanning, and the sampling position of each sampling moment is different; after the standard position is determined, the relative position of each sampling position relative to the standard position is obtained, the relative position is used for aligning the K space data to the standard position, the movable K space data are restored, and finally, the K space data are combined to obtain a magnetic resonance image without motion artifact, the K space data with large motion amplitude can be restored, and the artifact removal stability is excellent. Referring to fig. 1, fig. 1 is a flow chart of a magnetic resonance image acquisition method according to an embodiment of the present application, including:
S110, acquiring sampling positions and K space data of each sampling moment of a magnetic resonance scanning part of a detected person, wherein the sampling positions comprise coordinate data, pitch angle, yaw angle and roll angle;
When acquiring a magnetic resonance image, tens of thousands of pulses are required to acquire the image with enough definition, and K-space data and sampling positions at a plurality of moments are required to be acquired.
The K space data and the sampling position correspond to the magnetic resonance scanning position of the testee, the magnetic resonance scanning time is long, the magnetic resonance scanning position can be displaced, and the sampling position is changed. The sampling position comprises coordinate data, a pitch angle, a yaw angle and a roll angle; the coordinate data can be in the form of a rectangular coordinate system or in the form of polar coordinates; the pitch angle, yaw angle, roll angle in coordinates are used to accurately describe the pose of the patient in 3D space. The present embodiment is not limited to the subject, and may be a human or animal, and the magnetic resonance scanning portion may be a leg or brain.
S120, determining the relative position of each sampling position relative to the standard position;
in this embodiment, a standard position is determined, and a sampling position is obtained from an actual position directly acquired at a sampling time, then a difference value of each sampling position relative to the standard position is determined, and the difference value is determined as a relative position for aligning K space data. For example, a first time (coordinate data 1, pitch angle 1, yaw angle 1, roll angle 1) is acquired as a sampling position 1, a second time (coordinate data 2, pitch angle 2, yaw angle 2, roll angle 2) is acquired as a sampling position 2, and the standard positions are (coordinate data a, pitch angle a, yaw angle a, roll angle a); the relative position at the first moment can be (coordinate data 1-coordinate data a, pitch angle 1-pitch angle a, yaw angle 1-yaw angle a, roll angle 1-roll angle a); the relative position at the second time may be obtained as (coordinate data 2-coordinate data a, pitch angle 2-pitch angle a, yaw angle 2-yaw angle a, roll angle 2-roll angle a).
When the relative sampling position of the current sampling time to the last sampling time is acquired each time, the actual sampling position is required to be determined according to the relative sampling position, and then the actual sampling position is subjected to difference with the standard position, and the obtained difference value is the relative position. For example, the initial time acquisition (coordinate data s, pitch angle s, yaw angle s, roll angle s), the first time acquisition of the relative sampling position (coordinate data Δ1, pitch angle Δ1, yaw angle Δ1, roll angle Δ1) with respect to the initial time; the standard position is (coordinate data a, pitch angle a, yaw angle a, roll angle a); the relative positions of the initial moments are obtained as (coordinate data s-coordinate data a, pitch angle s-pitch angle a, yaw angle s-yaw angle a and roll angle s-roll angle a); the relative position at the first time is (coordinate data s+coordinate data Δ1-coordinate data a, pitch angle s+pitch angle Δ1-pitch angle a, yaw angle s+yaw angle Δ1-yaw angle a, roll angle Δ1+roll angle s-roll angle a).
S130, aligning the corresponding K space data according to the relative positions to obtain aligned K space data;
And carrying out space conversion on the K space data according to the relative positions, and aligning all the K space data at the standard positions to obtain aligned space data. For example, when the relative position is (coordinate data P, pitch angle P, yaw angle P, roll angle P), the K-space data is controlled to move horizontally-coordinate data P, roll-pitch angle P, -yaw angle P, -roll angle P.
S140, combining all the aligned K space data to obtain combined K space data;
s150, performing inverse Fourier transform on the combined K space data to obtain a magnetic resonance image of a space domain.
And combining all the aligned K space data, and performing inverse Fourier transform on the combined K space data to obtain a magnetic resonance image of a space domain.
Based on the above technical scheme, in the process of magnetic resonance scanning, each sampling time is used for acquiring K space data and simultaneously acquiring the sampling position of the magnetic resonance scanning position, the magnetic resonance scanning position of the subject can fluctuate during magnetic resonance scanning, and the sampling position of each sampling time is different; after the standard position is determined, the relative position of each sampling position relative to the standard position is obtained, the relative position is used for aligning the K space data to the standard position, the movable K space data are restored, and finally, the K space data are combined to obtain a magnetic resonance image without motion artifact, the K space data with large motion amplitude can be restored, and the artifact removal stability is excellent.
Based on the above embodiment, in order to improve the flexibility of determining the standard position, before determining the relative position of each sampling position with respect to the standard position, the method further includes: the standard position is determined from all the sampling positions, or the set position is taken as the standard position.
The standard position is selected from all the sampling positions according to a set rule, wherein the set rule can be any one of a designated initial position serving as the standard position, a designated end position serving as the standard position, a position corresponding to the middle sampling time serving as the standard position, an average position determined by all the sampling positions and a position kept longest. Furthermore, the set position is used as a standard position, and the set position can correspond to any fixed position on the magnetic resonance machine and can be selected according to actual requirements.
Based on the above technical scheme, the embodiment determines the standard position from all the sampling positions, or sets a fixed sampling position as the standard position, which can be flexibly set according to actual requirements, and improves user experience.
Based on any of the above embodiments, to reduce the introduction of errors, determining the standard position from all the sampling positions includes:
Determining a sampling position corresponding to the scanning of the middle sampling time in all the sampling times as a standard position;
or, determining the target sampling position with the largest number of times from all the sampling positions, and determining the target sampling position as the standard position.
The sampling position corresponding to any sampling time can be used as a standard position, but the sampling position closer to the middle sampling time or with smaller displacement in a continuous period of time is more suitable as the standard position mainly because data loss is inevitably caused when the rotation and translation operation of K space data is carried out, so that the sampling position is selected to be aligned at a position with a more approximate sequence as much as possible, and the introduction of movement and inspection errors is reduced.
Based on the above technical solution, in this embodiment, since the situation that there is data loss occurs when the K-space data is spatially converted, the sampling position corresponding to the intermediate sampling time or the target sampling position with the largest number of times determined in all the sampling positions is used as the standard position, so that movement can be reduced as much as possible, so as to reduce error introduction.
Based on any of the above embodiments, performing data alignment on the corresponding K-space data according to the relative position to obtain aligned K-space data, including: determining geometric transformation information according to the relative position, wherein the geometric transformation information comprises translation information and rotation information; and carrying out data transformation on the K space data by utilizing Fourier transformation according to the geometric transformation information to obtain transformed K space data, and taking the transformed K space data as aligned space data. Through the technical scheme, after translation information and rotation information are determined, data transformation is carried out on K space data by utilizing Fourier transformation, data aligned with a standard position is obtained, and the data is used as aligned space data.
Based on any of the above embodiments, further comprising: the sampling frequency of the magnetic resonance scan is controlled to be one half of the interval of two pulses. Tens of thousands of pulses are required to acquire a magnetic resonance image with sufficient definition. According to the nyquist sampling theory, the sampling frequency should ideally be close to half the interval of the two pulses when MR imaging. And because magnetic resonance imaging is interfered by magnetic metal, the current distance measurement technology may not meet the requirements of the design. The sampling frequency should be increased as much as possible during application to improve the accuracy of the simulation track.
Based on any of the above embodiments, in order to improve image quality of the magnetic resonance image, the magnetic resonance image acquisition method further includes: and inputting the magnetic resonance image into a pre-established neural network model for high-quality conversion to obtain a high-quality magnetic resonance image.
The process for establishing the neural network model comprises the following steps:
Acquiring a low-quality magnetic resonance image set and a corresponding high-quality magnetic resonance image set used during training;
inputting the low-quality magnetic resonance image set and the corresponding high-quality magnetic resonance image set into a preset neural network module for training to obtain an initial neural network model;
And verifying the initial neural network model by using the verification image, and determining the initial neural network model as a neural network model when the verification passes, wherein the neural network model can be practically used.
Wherein validating the initial neural network model with the validation image may include: inputting the verification image into an initial neural network model, outputting a high-quality verification image, displaying the high-quality verification image on a screen, evaluating the image by medical staff, transmitting an evaluation result to electronic equipment by the medical staff through client equipment, receiving the evaluation result by the electronic equipment, and determining that the verification is passed when the evaluation result is passed.
Furthermore, the neural network module is an antagonistic neural network module or a convolutional neural network module, and the user can set the neural network module according to actual requirements, so long as the purpose of the embodiment can be achieved.
According to the embodiment, a neural network model is obtained by training a low-quality medical image set and a high-quality medical image set, when a test image is processed through the neural network model, a diagnostician verifies according to the processed image, if the verification is passed, verification passing information is issued through equipment, at the moment, the neural network model is determined to be practically applicable, then the neural network model is utilized to process an image to be processed, and a high-quality image to be processed is output. The application converts the rapidly scanned low-quality image into the high-quality image through the neural network model, thereby realizing the effects of enhancing the signal-to-noise ratio, improving the resolution and reducing the artifact, and having low cost and conversion efficiency.
Based on the technical scheme, the neural network model for converting the low-quality image into the high-quality image is trained in advance, the magnetic resonance image is input into the neural network model, the high-quality magnetic resonance image is obtained, and the resolution of the image is improved.
Based on any of the above embodiments, acquiring the sampling position and K-space data for each sampling instant of the magnetic resonance scan region of the subject includes:
acquiring an initial sampling position and initial K space data of an initial sampling moment of a magnetic resonance scanning part of a detected person;
Acquiring a relative sampling position and corresponding current K space data of each sampling time of a magnetic resonance scanning part relative to the last sampling time;
after all the relative sampling positions are obtained, all the relative sampling positions are transformed by using the initial sampling positions to obtain all the sampling positions, and all the K space data are obtained according to the initial K space data and all the current K space data.
Based on the above technical scheme, the relative sampling positions based on the previous sampling time are collected each time, so that the data acquisition mode of the sampling time is simplified, and then after the sampling is completed, all the relative sampling positions are transformed by uniformly utilizing the initial sampling positions, so that all the sampling positions are obtained, and the efficiency of acquiring the final sampling positions is improved.
To further describe the description, the embodiment provides a specific method for acquiring magnetic resonance images, taking three acquisition moments as an example, including:
first, acquiring sampling positions and K space data at time t0, time t1 and time t2 during magnetic resonance scanning.
The sample position is described:
At time t0, the patient is exactly at the initial position of the rectangular coordinate system (x=0, y=0, z=0) at the start of radiography and the posture is also positive (pitch=0, yaw=0, roll=0). At this time, the sampling position is denoted by (0, 0) and recorded as data_0. The numbers of the guessing positions are (x, y, z, pitch, yaw, roll) respectively.
The patient turned 45 ° (roll=45°) to the right at time t1 (t=1), and also shifted 1 unit (x=2) to the left, the sampling position at this time being denoted by (2,0,0,0,0,45 °), and recorded as data_1.
The patient was raised by 60 ° (pitch=60°) at time t2 (t=2), and at the same time was also moved up by 1.7 units (z=1.7) and moved forward by 1 unit (y=1) on the basis of time t1, and the sampling position at this time was represented by (2,1,1.7,60 °,0,45 °), and was recorded as data_2.
Second, a standard position is determined from the sampling positions.
If the sampling position corresponding to t0 is taken as the standard position, the data alignment needs the following operations: data_0 is motionless; data_1 turns 45 ° to the left; move 1 unit to the right; data_2 is rotated back 60 °; move 1 unit backward; move 1 unit to the right; move 1.7 units downward; turn 45 to the left. The rotation operation is performed 3 times, and a total rotation operation is 150 °; the translation operation was performed a total of 4 times, 4.7 units.
If the sampling position corresponding to t1 is taken as the standard position, the operation required by data alignment is as follows: data_0 turns 45 ° to the right; move 1 unit to the left; data_1 is motionless; data_2 is rotated back 60 °; moves 1 unit backward. The rotation operation was performed 2 times, a total rotation operation of 105 °; the translation operation was performed 3 times in total, 3.7 units.
If the sampling position corresponding to t2 is taken as the standard position, the data alignment needs the following operations: data_0 turns 45 ° to the right; move 1 unit to the left; turning upwards by 60 degrees; move up 1.7 units; forward by 1 unit; data_1 is turned up 60 °; move up 1.7 units; forward by 1 unit; data_2 is stationary. The rotation operation is performed 3 times, a total rotation operation 165 °; the translation operation was performed a total of 5 times, 6.4 units.
The sampling position corresponding to t1 operates less as a standard position, introducing less error and thus selecting the sampling position corresponding to t 1.
And thirdly, determining the relative position.
Taking data_1 as a standard position, the relative position of data_0 is:
Data_0-Data_1=(0,0,0,0,0,0)-(2,0,0,0,0,45°)=(-2,0,0,0,0,-45°)。
the relative positions of data_2 are:
Data_2-Data_1=(2,1,1.7,60°,0,45°)-(2,0,0,0,0,45°)=(0,1,1.7,60°,0,0)。
and fourthly, aligning the data of the corresponding K space data according to the relative positions to obtain aligned K space data.
Carrying out space transformation on the K space data at the time t0 according to (-2,0,0,0,0, -45 DEG), and obtaining aligned K space data at the time t 0;
And carrying out space transformation on the K space data at the time t1 according to (0,1,1.7,60 degrees, 0) to obtain aligned K space data at the time t 1.
Fifthly, combining all the aligned K space data to obtain combined K space data; and performing inverse Fourier transform on the combined K space data to obtain a magnetic resonance image of the space domain.
The following describes a magnetic resonance image acquisition device according to an embodiment of the present application, and the magnetic resonance image acquisition device described below and the magnetic resonance image acquisition method described above can be referred to correspondingly, please refer to fig. 2, fig. 2 is a schematic structural diagram of the magnetic resonance image acquisition device according to the embodiment of the present application, which includes:
an acquisition module 210, configured to acquire sampling positions and K-space data of each sampling time of a magnetic resonance scanning position of a subject, where the sampling positions include coordinate data, pitch angle, yaw angle, and roll angle;
a relative position determining module 220, configured to determine a relative position of each sampling position with respect to the standard position;
an alignment module 230, configured to perform data alignment on the corresponding K-space data according to the relative position, so as to obtain aligned K-space data;
The merging module 240 is configured to merge all the aligned K-space data to obtain merged K-space data;
the fourier transform module 250 is configured to perform inverse fourier transform on the combined K-space data to obtain a magnetic resonance image in a spatial domain.
Based on the above technical scheme, in the process of magnetic resonance scanning, each sampling time is used for acquiring K space data and simultaneously acquiring the sampling position of the magnetic resonance scanning position, the magnetic resonance scanning position of the subject can fluctuate during magnetic resonance scanning, and the sampling position of each sampling time is different; after the standard position is determined, the relative position of each sampling position relative to the standard position is obtained, the relative position is used for aligning the K space data to the standard position, the movable K space data are restored, and finally, the K space data are combined to obtain a magnetic resonance image without motion artifact, the K space data with large motion amplitude can be restored, and the artifact removal stability is excellent.
Preferably, the method further comprises:
and the standard position determining module is used for determining the standard position from all sampling positions or taking the set position as the standard position.
Preferably, the standard position determining module includes:
the first standard position determining unit is used for acquiring sampling positions and K space data of each sampling moment of a magnetic resonance scanning position of a detected person, wherein the sampling positions comprise coordinate data, pitch angle, yaw angle and rolling angle;
And the second standard position determining unit is used for determining the target sampling position with the largest number of times from all the sampling positions and determining the target sampling position as the standard position.
Preferably, the alignment module 230 includes:
a geometric transformation information determining unit for determining geometric transformation information according to the relative position, wherein the geometric transformation information includes translation information and rotation information;
And the alignment unit is used for carrying out data transformation on the K space data by utilizing Fourier transformation according to the geometric transformation information to obtain transformed K space data, and taking the transformed K space data as aligned space data.
Preferably, the method further comprises:
and the sampling frequency control module is used for controlling the sampling frequency of the magnetic resonance scanning to be one half of the interval of two pulses.
Preferably, the method further comprises:
the high-quality conversion module is used for inputting the magnetic resonance image into a pre-established neural network model to perform high-quality conversion, so as to obtain the high-quality magnetic resonance image.
Preferably, the acquisition module 210 includes:
An initial acquisition unit for acquiring initial sampling position and initial K space data of an initial sampling time of a magnetic resonance scanning part of a subject;
the second unit is used for acquiring the relative sampling position of each sampling time of the magnetic resonance scanning part relative to the last sampling time and corresponding current K space data;
And the acquisition unit is used for transforming all the relative sampling positions by using the initial sampling positions after acquiring all the relative sampling positions to obtain all the sampling positions, and obtaining all the K space data according to the initial K space data and all the current K space data.
The following describes an electronic device provided in an embodiment of the present application, where the electronic device described below and the magnetic resonance image acquisition method described above may be referred to correspondingly.
In an embodiment of the present application, as shown in fig. 3, an electronic device 300 shown in fig. 3 includes: a processor 301 and a memory 303. Wherein the processor 301 is coupled to the memory 303, such as via a bus 302. Optionally, the electronic device 300 may also include a transceiver 304. It should be noted that, in practical applications, the transceiver 304 is not limited to one, and the structure of the electronic device 300 is not limited to the embodiment of the present application.
Based on the above technical scheme, in the process of magnetic resonance scanning, each sampling time is used for acquiring K space data and simultaneously acquiring the sampling position of the magnetic resonance scanning position, the magnetic resonance scanning position of the subject can fluctuate during magnetic resonance scanning, and the sampling position of each sampling time is different; after the standard position is determined, the relative position of each sampling position relative to the standard position is obtained, the relative position is used for aligning the K space data to the standard position, the movable K space data are restored, and finally, the K space data are combined to obtain a magnetic resonance image without motion artifact, the K space data with large motion amplitude can be restored, and the artifact removal stability is excellent.
The Processor 301 may be a CPU (Central Processing Unit ), general purpose Processor, DSP (DIGITAL SIGNAL Processor, data signal Processor), ASIC (Application SPECIFIC INTEGRATED Circuit), FPGA (Field Programmable GATE ARRAY ) or other programmable logic device, transistor logic device, hardware component, or any combination thereof. Which may implement or perform the various exemplary logic blocks, modules and circuits described in connection with this disclosure. Processor 301 may also be a combination that implements computing functionality, e.g., comprising one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.
Bus 302 may include a path to transfer information between the components. Bus 302 may be a PCI (PERIPHERAL COMPONENT INTERCONNECT, peripheral component interconnect standard) bus or an EISA (Extended Industry Standard Architecture ) bus, or the like. Bus 302 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in fig. 3, but not only one bus or one type of bus.
The Memory 303 may be, but is not limited to, a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory ) or other type of dynamic storage device that can store information and instructions, an EEPROM (ELECTRICALLY ERASABLE PROGRAMMABLE READ ONLY MEMORY ), a CD-ROM (Compact Disc Read Only Memory, compact disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer.
The memory 303 is used for storing application program codes for executing the inventive arrangements and is controlled to be executed by the processor 301. The processor 301 is configured to execute the application code stored in the memory 303 to implement what is shown in the foregoing method embodiments.
Among them, electronic devices include, but are not limited to: mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and stationary terminals such as digital TVs, desktop computers, and the like. The electronic device shown in fig. 3 is only an example and should not be construed as limiting the functionality and scope of use of the embodiments of the application.
A computer readable storage medium provided in the embodiments of the present application is described below, and the computer readable storage medium described below and the method described above may be referred to correspondingly.
A computer readable storage medium having instructions stored therein which, when executed on a computer, cause the computer to perform a magnetic resonance image acquisition method as above.
Based on the above technical scheme, in the process of magnetic resonance scanning, each sampling time is used for acquiring K space data and simultaneously acquiring the sampling position of the magnetic resonance scanning position, the magnetic resonance scanning position of the subject can fluctuate during magnetic resonance scanning, and the sampling position of each sampling time is different; after the standard position is determined, the relative position of each sampling position relative to the standard position is obtained, the relative position is used for aligning the K space data to the standard position, the movable K space data are restored, and finally, the K space data are combined to obtain a magnetic resonance image without motion artifact, the K space data with large motion amplitude can be restored, and the artifact removal stability is excellent.
It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated herein. Moreover, at least some of the steps in the flowcharts of the figures may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily being sequential, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.
The foregoing is only a partial embodiment of the present application, and it should be noted that it will be apparent to those skilled in the art that modifications and adaptations can be made without departing from the principles of the present application, and such modifications and adaptations should and are intended to be comprehended within the scope of the present application.
Claims (10)
1. A method of acquiring a magnetic resonance image, comprising:
acquiring sampling positions and K space data of each sampling moment of a magnetic resonance scanning part of a detected person, wherein the sampling positions comprise coordinate data, pitch angle, yaw angle and rolling angle;
determining the relative position of each sampling position relative to a standard position;
Data alignment is carried out on the corresponding K space data according to the relative positions, and K space data after alignment is obtained;
combining all the aligned K space data to obtain combined K space data;
and performing inverse Fourier transform on the combined K space data to obtain a magnetic resonance image of a space domain.
2. The method of claim 1, wherein prior to determining the relative position of each of the sampling locations with respect to a standard location, further comprising:
And determining the standard position from all the sampling positions, or taking a set position as the standard position.
3. The method of claim 1, wherein said determining the standard position from all of the sample positions comprises:
Determining the sampling position corresponding to the scanning of the middle sampling time in all the sampling times as the standard position;
Or determining the target sampling position with the same number of times and the maximum number of times from all the sampling positions, and determining the target sampling position as the standard position.
4. The method of claim 1, wherein the performing data alignment on the corresponding K-space data according to the relative position to obtain aligned K-space data includes:
determining geometric transformation information according to the relative position, wherein the geometric transformation information comprises translation information and rotation information;
and carrying out data transformation on the K space data by utilizing Fourier transformation according to the geometric transformation information to obtain transformed K space data, and taking the transformed K space data as the aligned space data.
5. The method of magnetic resonance image acquisition as set forth in claim 1, further comprising:
the sampling frequency of the magnetic resonance scan is controlled to be one half of the interval of two pulses.
6. The method of magnetic resonance image acquisition as set forth in claim 1, further comprising:
And inputting the magnetic resonance image into a pre-established neural network model for high-quality conversion to obtain a high-quality magnetic resonance image.
7. The method of any one of claims 1 to 6, wherein acquiring the sampling position and K-space data for each sampling instant of the magnetic resonance scan region of the subject comprises:
acquiring initial sampling positions and initial K space data of initial sampling moments of the magnetic resonance scanning part of the subject;
Acquiring a relative sampling position and corresponding current K space data of each sampling time of the magnetic resonance scanning part relative to the last sampling time;
And after all the relative sampling positions are acquired, transforming all the relative sampling positions by utilizing the initial sampling positions to obtain all the sampling positions, and obtaining all the K space data according to the initial K space data and all the current K space data.
8. A magnetic resonance image acquisition apparatus, comprising:
The acquisition module is used for acquiring sampling positions and K space data of each sampling moment of a magnetic resonance scanning part of a detected person, wherein the sampling positions comprise coordinate data, pitch angle, yaw angle and rolling angle;
The relative position determining module is used for determining the relative position of each sampling position relative to the standard position;
The alignment module is used for carrying out data alignment on the corresponding K space data according to the relative position to obtain aligned K space data;
the merging module is used for merging all the K-space data after alignment to obtain merged K-space data;
And the Fourier transform module is used for carrying out inverse Fourier transform on the combined K space data to obtain a magnetic resonance image of a space domain.
9. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program capable of being loaded by the processor and performing the magnetic resonance image acquisition method according to any one of claims 1 to 7.
10. A computer readable storage medium having instructions stored therein, which when executed on a computer, cause the computer to perform the magnetic resonance image acquisition method as claimed in any one of claims 1 to 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111428699.8A CN114266728B (en) | 2021-11-27 | 2021-11-27 | Method, device, equipment and medium for acquiring magnetic resonance image |
| PCT/CN2021/138541 WO2023092742A1 (en) | 2021-11-27 | 2021-12-15 | Method and apparatus for acquiring magnetic resonance image, and device and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111428699.8A CN114266728B (en) | 2021-11-27 | 2021-11-27 | Method, device, equipment and medium for acquiring magnetic resonance image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114266728A CN114266728A (en) | 2022-04-01 |
| CN114266728B true CN114266728B (en) | 2024-08-23 |
Family
ID=80825668
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111428699.8A Active CN114266728B (en) | 2021-11-27 | 2021-11-27 | Method, device, equipment and medium for acquiring magnetic resonance image |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114266728B (en) |
| WO (1) | WO2023092742A1 (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110664378A (en) * | 2019-10-28 | 2020-01-10 | 中国科学院深圳先进技术研究院 | Magnetic resonance imaging method, device, system and storage medium |
| CN111443318A (en) * | 2019-01-16 | 2020-07-24 | 上海联影医疗科技有限公司 | Magnetic resonance image processing method, magnetic resonance image processing device, storage medium and magnetic resonance imaging system |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9535146B2 (en) * | 2011-09-29 | 2017-01-03 | Hitachi, Ltd. | Magnetic resonance imaging apparatus and magnetic resonance imaging method |
| US10012717B2 (en) * | 2015-04-14 | 2018-07-03 | Siemens Healthcare Gmbh | Spatially varying coil compression in magnetic resonance image reconstruction |
| EP3447518B1 (en) * | 2017-08-25 | 2022-01-26 | Siemens Healthcare GmbH | Fan- or cone-beam geometry for magnetic resonance imaging and x-ray imaging |
| JP7509546B2 (en) * | 2020-02-12 | 2024-07-02 | キヤノンメディカルシステムズ株式会社 | Image Reconstruction Device |
| KR102144320B1 (en) * | 2020-02-13 | 2020-08-13 | 주식회사 에어스 메디컬 | Magnetic resonance image processing apparatus and method |
| CN112557980B (en) * | 2020-11-02 | 2022-05-03 | 上海东软医疗科技有限公司 | Magnetic resonance image correction method, magnetic resonance image correction device, medium, and electronic apparatus |
| CN113516603B (en) * | 2021-08-03 | 2022-08-19 | 上海交通大学 | Magnetic resonance image multi-modal reconstruction method based on spatial fusion and storage medium |
-
2021
- 2021-11-27 CN CN202111428699.8A patent/CN114266728B/en active Active
- 2021-12-15 WO PCT/CN2021/138541 patent/WO2023092742A1/en unknown
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111443318A (en) * | 2019-01-16 | 2020-07-24 | 上海联影医疗科技有限公司 | Magnetic resonance image processing method, magnetic resonance image processing device, storage medium and magnetic resonance imaging system |
| CN110664378A (en) * | 2019-10-28 | 2020-01-10 | 中国科学院深圳先进技术研究院 | Magnetic resonance imaging method, device, system and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114266728A (en) | 2022-04-01 |
| WO2023092742A1 (en) | 2023-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Johnson et al. | Conditional generative adversarial network for 3D rigid‐body motion correction in MRI | |
| Küstner et al. | Retrospective correction of motion‐affected MR images using deep learning frameworks | |
| CN101232845B (en) | Magnetic resonance imaging device | |
| Forman et al. | Self-encoded marker for optical prospective head motion correction in MRI | |
| US20170276754A1 (en) | Methods, systems, and devices for intra-scan motion correction | |
| CN103260510B (en) | MR imaging apparatus and contrast strengthen image acquisition method | |
| US20140055135A1 (en) | Motion tracking based on fast image acquisition | |
| US11255943B2 (en) | Motion determination for volumetric magnetic resonance imaging using a deep machine-learning model | |
| CN107037386B (en) | A kind of echo planar imaging and system | |
| Liu et al. | Learning MRI artefact removal with unpaired data | |
| US11514622B2 (en) | System, apparatus, and method for incremental motion correction in magnetic resonance imaging | |
| US20230401691A1 (en) | Image defect detection method, electronic device and readable storage medium | |
| CN111047660A (en) | Image reconstruction method, device, equipment and storage medium | |
| Zou et al. | Multi-scale deformable transformer for multi-contrast knee MRI super-resolution | |
| CN111257809A (en) | Magnetic resonance imaging method, magnetic resonance imaging device, storage medium and medical equipment | |
| CN111009032A (en) | Blood vessel three-dimensional reconstruction method based on improved epipolar line constraint matching | |
| WO2019167597A1 (en) | Super-resolution processing device and method, and program | |
| CN114266728B (en) | Method, device, equipment and medium for acquiring magnetic resonance image | |
| CN114581549B (en) | Four-dimensional ultrasonic reconstruction method and system based on two-dimensional ultrasonic image | |
| Jaubert et al. | HyperSLICE: HyperBand optimized spiral for low‐latency interactive cardiac examination | |
| JP6781417B2 (en) | Diencephalon registration | |
| CN111161330B (en) | Non-rigid image registration method, device, system, electronic equipment and storage medium | |
| CN104173049A (en) | Method for recording image data sets with a magnetic resonance tomography system | |
| CN112700439A (en) | OCT human eye image acquisition and registration method and system based on neural network | |
| JP2014203162A (en) | Inclination angle estimation device, mtf measuring apparatus, inclination angle estimation program and mtf measurement program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |