CN112288687B - Inner ear space posture analysis method and analysis system - Google Patents
Inner ear space posture analysis method and analysis system Download PDFInfo
- Publication number
- CN112288687B CN112288687B CN202010935021.8A CN202010935021A CN112288687B CN 112288687 B CN112288687 B CN 112288687B CN 202010935021 A CN202010935021 A CN 202010935021A CN 112288687 B CN112288687 B CN 112288687B
- Authority
- CN
- China
- Prior art keywords
- semicircular canal
- inner ear
- semicircular
- canal
- plane
- 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
- 210000003027 ear inner Anatomy 0.000 title claims abstract description 101
- 238000004458 analytical method Methods 0.000 title claims abstract description 41
- 210000002480 semicircular canal Anatomy 0.000 claims abstract description 257
- 239000013598 vector Substances 0.000 claims abstract description 73
- 210000003582 temporal bone Anatomy 0.000 claims abstract description 64
- 238000013178 mathematical model Methods 0.000 claims abstract description 51
- 210000005252 bulbus oculi Anatomy 0.000 claims abstract description 50
- 238000000034 method Methods 0.000 claims abstract description 50
- 238000003745 diagnosis Methods 0.000 claims abstract description 34
- 238000011282 treatment Methods 0.000 claims abstract description 34
- 230000011218 segmentation Effects 0.000 claims abstract description 30
- 210000003477 cochlea Anatomy 0.000 claims abstract description 11
- 230000002146 bilateral effect Effects 0.000 claims description 22
- 210000003128 head Anatomy 0.000 claims description 20
- 238000010276 construction Methods 0.000 claims description 18
- 230000008569 process Effects 0.000 claims description 7
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 238000003491 array Methods 0.000 claims description 3
- 206010047348 Vertigo positional Diseases 0.000 abstract description 16
- 208000001870 benign paroxysmal positional vertigo Diseases 0.000 abstract description 16
- 238000012360 testing method Methods 0.000 description 10
- 241000356847 Otolithes Species 0.000 description 6
- 230000002842 otolith Effects 0.000 description 6
- 210000001265 otolithic membrane Anatomy 0.000 description 6
- 230000009286 beneficial effect Effects 0.000 description 5
- 239000003708 ampul Substances 0.000 description 4
- 230000005484 gravity Effects 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 241000656145 Thyrsites atun Species 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 210000001508 eye Anatomy 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000007792 addition Methods 0.000 description 2
- 201000000691 benign paroxysmal positional nystagmus Diseases 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 208000002173 dizziness Diseases 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005096 rolling process Methods 0.000 description 2
- 208000027601 Inner ear disease Diseases 0.000 description 1
- 238000005481 NMR spectroscopy Methods 0.000 description 1
- 206010069689 Spinal column injury Diseases 0.000 description 1
- 208000012886 Vertigo Diseases 0.000 description 1
- 210000003484 anatomy Anatomy 0.000 description 1
- 210000000467 autonomic pathway Anatomy 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 238000000750 constant-initial-state spectroscopy Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002964 excitative effect Effects 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000035876 healing Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000013421 nuclear magnetic resonance imaging Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000000638 stimulation Effects 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 230000002792 vascular Effects 0.000 description 1
- 231100000889 vertigo Toxicity 0.000 description 1
- 230000001720 vestibular Effects 0.000 description 1
- 208000027491 vestibular disease Diseases 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30041—Eye; Retina; Ophthalmic
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Theoretical Computer Science (AREA)
- Epidemiology (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Primary Health Care (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Eye Examination Apparatus (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an inner ear space attitude analysis method and an inner ear space attitude analysis system, wherein the method comprises the following steps: model segmentation is carried out on the outer semicircular canal, the rear semicircular canal, the upper semicircular canal, the main pipe, the vestibule, the cochlea and the eyeball structure in the temporal bone image; establishing a horizontal plane to construct an inner ear space coordinate system by using a semicircular canal fundus plane in the segmented temporal bone image; calculating unit vectors of plane normal vectors of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal in a space coordinate system, constructing a human body semicircular canal space posture mathematical model, analyzing temporal bone image data of the BBPV patient by using the method, and making an individualized BBPV diagnosis and treatment scheme of the patient. The advantages are as follows: the automatic real-time segmentation can be realized, and the clinical application is facilitated; through the inner ear space three-dimensional coordinate system, the space position of the semicircular canal can be conveniently determined, and then the semicircular canal posture mathematical model is constructed, so that the method is directly used for making the accurate individuation BPPV diagnosis and treatment scheme of the patient, and the diagnosis and treatment operation is intuitively guided.
Description
[ field of technology ]
The invention relates to the technical field of medical image processing, in particular to an inner ear space posture analysis method and an inner ear space posture analysis system.
[ background Art ]
Benign paroxysmal positional vertigo (Benign paroxysmal positional vertigo, BBPV) is a clinically common peripheral vestibular disorder, the most common vertigo from the inner ear. When the head moves to a specific position, short dizziness can be induced, eye shake and autonomic nerve symptoms are accompanied, the disease has self-limiting property, and the most commonly involved semicircular canals are a posterior semicircular canal, an external semicircular canal and an upper semicircular canal in sequence.
In clinic, otolith is induced to subside under the action of gravity in a specific semicircular canal through head position change, fluid mechanics acts on the ampulla crest, and dizziness and eye vibration are induced to judge the otolith position. The external semicircular BPPV is diagnosed using the Dix-Hallpike test, followed by the upper semicircular BPPV, horizontal roll test.
The displaced otolith can be restored to the elliptical sac again through a series of head position changes, so that the healing can be achieved. The Epley method is commonly used to reset the posterior semicircular canal BPPV and the tumbling reset method is used to reset the external semicircular canal BPPV.
Because of the large difference of the space directions of the individual semicircular canals, the individual diagnosis and treatment methods can be designed according to the space directions of the individual semicircular canals of the patients.
The study on the space direction of the semicircular canal is insufficient, and the low-head horizontal Dix-Hall pike test and the 60-degree horizontal rolling test are more reasonable in design based on the study on the space anatomy of the semicircular canal.
The position of the semicircular canal in the space coordinate system is a precondition for BPPV diagnosis and treatment operation and research on the spatial direction of the semicircular canal. The BBPV diagnosis and treatment scheme making process by analyzing the semicircular canal posture mainly comprises the following steps: based on the division of main structures (an outer semicircular canal, a rear semicircular canal, an upper semicircular canal, a main pipe, a vestibule, a cochlea and an eyeball) on the inner ear temporal bone image, the space direction of the inner ear is determined by establishing a space three-dimensional coordinate system, and an individual diagnosis and treatment method is designed by measuring and calculating the included angle between the semicircular canal and each coordinate plane in the space coordinate system. Common three-dimensional space coordinate systems include the frankfurt coordinate system and the Reid coordinate system. The frankfurt plane is a plane defined by three points, namely the upper points of the left and right eardoors and the lower edge point of the left orbit. The Reid plane is a plane determined by three points of the midpoint of the left and right auricle doors and the infraorbital rim.
Many studies have demonstrated that when a person takes an upright posture with both eyes looking forward on a plane, the frankfurt plane is parallel to the true horizontal plane, and that the coronal plane can be established based on symmetry of left and right semicircular canals; the horizontal plane can be established according to the frankfurt plane, but the automatic acquisition of the frankfurt plane by a computer is difficult.
[ invention ]
Aiming at the defects in the prior art, the invention provides an inner ear space gesture analysis method and an inner ear space gesture analysis system, which can more accurately divide and acquire a semicircular canal model through a 3D convolutional neural network, can realize automatic real-time division and are beneficial to clinical application; the inner ear space three-dimensional coordinate system is built by taking the plane formed by the top end of the common tube of the two-side semicircular canals and the bottom of the two-side eyeballs as a horizontal plane, so that the space position of the semicircular canals can be conveniently determined, further, a semicircular canal posture mathematical model is built, and the method is directly used for making the accurate BPPV diagnosis and treatment scheme of the patient and intuitively guiding diagnosis and treatment operation.
In order to achieve the above object, according to a first aspect, the present invention adopts the following technical scheme: an inner ear spatial pose analysis method, comprising the steps of:
constructing a segmentation model according to temporal bone image data, wherein the temporal bone structure segmented by the model comprises: outer semicircular canal, posterior semicircular canal, superior semicircular canal, main canal, vestibule, cochlea and eyeball;
the divided temporal bone image is rotated to enable the top end of a main pipe and the bottom of an eyeball to be positioned in the cross section, and the inner ears on two sides are symmetrical relative to the sagittal plane, so that a space coordinate system is constructed;
obtaining central lines of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal from the segmented temporal bone image, converting the central lines into point arrays, adopting a least square method to fit planes respectively to obtain plane equations of the semicircular canals, calculating unit vectors of plane normal vectors of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal in the space coordinate system, and constructing an individual semicircular canal space posture mathematical model;
detecting a plurality of double-sided inner ear models, calculating a unit vector of the sum of normal vectors of all semicircular canal planes of the models in the space coordinate system, using the unit vector as a mathematical model of human semicircular canal space posture to analyze temporal bone image data of a BBPV patient, and making an individualized BBPV diagnosis and treatment scheme of the patient.
Preferably, the method comprises the steps of,
the method is characterized in that a segmentation model is constructed according to temporal bone image data, and the method is realized by the following steps:
deriving temporal bone image sample data from a medical image information system, and storing the temporal bone image sample data in a Dicom format;
performing recognition training on an inner ear structure and an eyeball in temporal bone image sample data by adopting a 3D convolutional neural network, and constructing an automatic segmentation model of a semicircular canal and the eyeball;
the temporal bone image is automatically segmented through the automatic segmentation model, and the inner ear structure is segmented into: external semicircular canal, posterior semicircular canal, superior semicircular canal, total canal, vestibule, cochlea and eyeball.
Preferably, the method comprises the steps of,
the construction of the space coordinate system is realized by the following steps:
selecting a plane formed by the lowest edge of the bilateral eyeball and the top end of the bilateral semicircular canal in the segmented temporal bone image, wherein the plane is used as a horizontal plane of a space coordinate system, and the normal vector of the plane is the Z axis of the space coordinate system;
taking the connecting lines at the top ends of the manifolds at two sides as X axes, and taking the X and Z axes as Y axes.
Preferably, the method comprises the steps of,
the construction process of the Z axis of the space coordinate system comprises the following steps:
respectively acquiring central lines of each outer semicircular canal, each rear semicircular canal and each upper semicircular canal in the inner ear from the segmented temporal bone image;
acquiring the intersection point of a central ring formed by the central lines of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal, and converting the inner ear central line model into a point array;
calculating the connection distance of each point between the intersecting points, removing abnormal intersecting points according to the length of the central ring of each semicircular tube, and identifying the spatial positions of the outer semicircular tube, the rear semicircular tube and the upper semicircular tube, wherein the intersecting point above the rear semicircular tube and the upper semicircular tube is correspondingly the top end of a main pipe;
taking the lowest point of eyeball coordinates in the inner ear and the top ends of the main pipes on two sides to form a plane and calculating a plane equation;
and analyzing the distance between each point of the eyeball and the plane, and taking the point with the largest distance, wherein the point and the top ends of the main pipes on two sides form a semicircular canal eyeball plane as a cross section, and the normal vector of the cross section is a Z axis.
Preferably, the method comprises the steps of,
the inner ear space gesture analysis method further comprises the following steps: the method for constructing the individual semicircular canal space posture mathematical model comprises the steps of calculating to obtain a human body semicircular canal space posture mathematical model by detecting a plurality of groups of bilateral inner ear models, wherein the human body semicircular canal space posture mathematical model is composed of average unit normal vectors of all planes of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal, and the average unit normal vectors are obtained through the following steps:
obtaining at least 100 divided double-sided inner ear models, respectively calculating the direction angle of each semicircular tube plane in the double-sided inner ear models and each coordinate axis in the space coordinate system, and establishing a unit normal vector according to the cosine of the direction angle;
and calculating a unit vector of the vector sum of the unit normal vectors, wherein the unit vector is an average unit normal vector.
Preferably, the method comprises the steps of,
the inner ear space gesture analysis method further comprises the following steps: and analyzing temporal bone image data of the BBPV patient according to the human body semicircular canal space posture mathematical model, and making an individualized BBPV diagnosis and treatment scheme of the patient.
Preferably, the method comprises the steps of,
the personalized BBPV diagnosis and treatment scheme comprises:
calculating cosine average values of included angles between the rear semicircular canal and each coordinate axis by using the plane normal vector of the rear semicircular canal of the patient in the human body semicircular canal space posture mathematical model to obtain Euler angle schemes and rotation paths from the rotation of the rear semicircular canal to the sagittal plane parallelism, and/or
And calculating cosine average values of included angles between the outer semicircular canal and each coordinate axis by using the outer semicircular canal plane normal vector of the patient in the human body semicircular canal space posture mathematical model, and obtaining an Euler angle scheme and a rotating path of the outer semicircular canal rotating to the coronal plane parallel.
In order to achieve the above object, according to a second aspect of the present invention, there is provided a technical scheme that: an inner ear spatial pose analysis system comprising:
the segmentation module is used for carrying out model segmentation on the structures of the outer semicircular canal, the rear semicircular canal, the upper semicircular canal, the main pipe, the vestibule, the cochlea and the eyeball in the temporal bone image;
the space coordinate system construction module is used for establishing a horizontal plane to construct an inner ear space coordinate system by using the semicircular canal fundus plane in the segmented temporal bone image;
the model construction module is used for calculating unit vectors of plane normal vectors of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal in a space coordinate system and constructing a human body semicircular canal space posture mathematical model;
the scheme making module is used for calculating a human body semicircular canal space posture mathematical model, analyzing temporal bone image data of the BBPV patient and making an individualized BBPV diagnosis and treatment scheme of the patient.
Preferably, the method comprises the steps of,
the space coordinate system takes a plane normal vector formed by the lowest edge of the eyeball at the two sides of the head of the human body and the top end of the common tube of the two sides of the semicircular canal as a Z axis, takes the top end connecting line of the common tube of the two sides of the semicircular canal as an X axis, and takes the cross of the X axis and the Z axis as a Y axis.
Preferably, the method comprises the steps of,
the space coordinate system construction module comprises a conversion unit, wherein the conversion unit is used for obtaining the central line of each semicircular canal in the segmented temporal bone image, converting the inner ear central line model into a point array, and identifying the space position by analyzing the cross point of the central ring of each semicircular canal.
The invention has the advantages that:
1. according to the invention, the semicircular canal model can be obtained by constructing the bilateral inner ear segmentation model and more accurately segmenting through the 3D convolutional neural network, and meanwhile, the automatic real-time segmentation can be realized, so that the method is beneficial to clinical application.
2. According to the invention, the plane formed by the top end of the common tube of the two-side semicircular canals and the midpoint of the two-side eyeballs is used as a horizontal plane to construct the inner ear space three-dimensional coordinate system, so that the space position of the semicircular canals can be conveniently determined, further, a semicircular canal posture mathematical model is constructed, the accurate individual BPPV diagnosis and treatment scheme of the patient is directly generated, and the diagnosis and treatment operation is intuitively guided.
3. The method can be used for analyzing temporal bone image data of BBPV patients, and the individual semicircular canal space posture mathematical model generated by the method can be used for making an individual BBPV diagnosis and treatment method of the patients.
[ description of the drawings ]
In order to more clearly illustrate the embodiments of the invention or the prior art solutions, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some of the embodiments described in the present invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
FIG. 1 is a flow chart of the inner ear spatial pose analysis method of the present invention;
FIG. 2 is a schematic diagram of the framework of the inner ear spatial pose analysis system of the present invention.
[ detailed description ] of the invention
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to specific embodiments and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
According to the inner ear space attitude analysis method and the inner ear space attitude analysis system, a 3D convolutional neural network is adopted to divide and acquire a semicircular canal model to realize automatic division, an inner ear space three-dimensional coordinate system is built on a divided temporal bone image, the space positions of semicircular canals in the inner ear are identified by converting an inner ear central line model into a point group, the lowest point of eyeball coordinates and the top ends of two side main pipes are used as the horizontal planes of the space coordinate system to build a three-dimensional coordinate system, a least square method is adopted to fit the planes of the rear semicircular canal, the upper semicircular canal and the outer semicircular canal respectively to acquire plane equations of the three-dimensional coordinate system, the average unit normal vector of each semicircular canal plane is calculated to serve as a human body semicircular canal space attitude mathematical model, a precise BPPV diagnosis and treatment scheme of a patient is automatically generated through the human body semicircular canal space attitude mathematical model, and diagnosis and treatment operation is intuitively guided.
As shown in fig. 1, the inner ear space gesture analysis method of the present invention comprises the steps of:
s1, constructing a segmentation model according to temporal bone image data, wherein a temporal bone structure segmented by the model comprises: external semicircular canal, posterior semicircular canal, superior semicircular canal, total canal, vestibule, cochlea and eyeball. The temporal bone image in the invention can fully utilize the spatial information of the temporal bone scanning image through the 3D coding and decoding network, and realize the automatic segmentation of the inner ear and the eyeball of the temporal bone image. In the prior art, a computer tomography or nuclear magnetic resonance imaging is adopted to construct a human inner ear space virtual reality image, the inner ear space gesture is presented in real time based on the image, but bone marker points are difficult to obtain based on nuclear magnetic resonance.
S2, rotating the segmented temporal bone image to enable the top end of the main pipe and the bottom of the eyeball to be located in the cross section, and enabling the inner ears on two sides to be symmetrical relative to the sagittal plane, so that a space coordinate system is constructed. The biggest disadvantage of the semi-automatization research in the technical field of the existing inner ear space image is that a head space coordinate system is not established, and each established semicircular tube plane equation is not located in a standard space coordinate system, so that the application is limited, for example: the frankfurt coordinate system and the Reid coordinate system, the frankfurt plane is a plane determined by three points of upper points of left and right side eardoors and lower points of left side orbits, the Reid plane is a plane determined by three points of middle points of left and right side eardoors and lower points of orbits, and the head space coordinate system cannot be established according to the two planes because mark points are difficult to determine or the scanning range is insufficient. The invention establishes the head space coordinate system by determining the plane formed by the top end of the semicircular canal main pipe and the eyeball bottom and taking the plane as the cross section, thereby being beneficial to realizing machine learning, further realizing automatic construction and adjustment of the standard space coordinate system and establishing the semicircular canal gesture mathematical model.
S3, obtaining central lines of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal from the segmented temporal bone image, converting the central lines into point arrays, adopting a least square method to fit planes respectively to obtain plane equations of the central lines, calculating unit vectors of plane normal vectors of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal in the space coordinate system, and constructing an individual semicircular canal space posture mathematical model. The process of constructing the individual semicircular canal space posture mathematical model through the steps S1-S3 can analyze temporal bone image data of any BBPV patient, and the individual semicircular canal space posture mathematical model corresponding to the patient is formulated, and the personalized BBPV diagnosis and treatment scheme of the patient can be formulated through the model.
S4, detecting a plurality of cases of bilateral inner ear models, calculating a unit vector of the sum of normal vectors of semicircular canal planes of the models in the space coordinate system, using the unit vector as a human body semicircular canal space posture mathematical model to analyze temporal bone image data of the BBPV patient, and making an individualized BBPV diagnosis and treatment scheme of the patient. The analysis of temporal bone image data of the BBPV patient is essentially a process of constructing an individual semicircular canal space posture model in the steps 21-S3, the complete analysis of temporal bone image data of the BBPV patient is carried out through the process, and then an individual BBPV diagnosis and treatment scheme of the patient is formulated through a human semicircular canal space posture mathematical model obtained through calculation of a plurality of cases of bilateral inner ear models. The personalized BBPV diagnosis and treatment scheme comprises: and calculating an included angle cosine average value between the rear semicircular canal and each coordinate axis by using the rear semicircular canal plane normal vector of the patient in the human body semicircular canal space posture mathematical model to obtain an Euler angle scheme and a rotating path of the rear semicircular canal rotating to the sagittal plane parallel, and/or calculating an included angle cosine average value between the outer semicircular canal and each coordinate axis by using the outer semicircular canal plane normal vector of the patient in the human body semicircular canal space posture mathematical model to obtain an Euler angle scheme and a rotating path of the outer semicircular canal rotating to the coronal plane parallel.
Through the mathematical model of the space posture of the human semicircular canal, the temporal bone image of the patient can be automatically fitted to a plane, the average space direction of the semicircular canal is calculated, and the automatic analysis of the space posture of the inner ear of the patient is facilitated.
In a preferred embodiment, the step S1 of the present invention is implemented by constructing a bilateral inner ear segmentation model according to temporal bone image data, specifically by the following steps:
s11, deriving original temporal bone image sample data from a medical image information system (Picture Archiving and Communication Systems, PACS), and storing the data into a Dicom format (Digital Imaging and Communications in Medicine, DICOM, digital imaging and communication in medical science, which is an international standard (ISO 12052) of medical images and related information, and specifically represents a medical image format which can be used for data exchange and has quality meeting clinical requirements); the catalog is read by 3D Slicer4.10.2 software to obtain images, and the 3D-CISS sequence is automatically exported and stored into NII format.
S12, performing recognition training on an inner ear structure and an eyeball in the temporal bone image sample data by using a 3D convolutional neural network, and constructing an automatic segmentation model of the semicircular canal and the eyeball. Specifically, the invention is based on a 3D Unet network, trains inner ear structures and eyeball identifications in a plurality of samples, uses ResNet as a BackBone model (backBone) to replace the original DoubleConv, and establishes an automatic segmentation model of a semicircular canal and an eyeball. The 3D UNet downsampled BackNone is stacked by common CBR modules (Conv+BN+ReLU), and the ResNet which is more learned is regarded as the BackNone of the 3D UNet, so that the segmentation effect is better.
S13, automatically segmenting the temporal bone image through the automatic segmentation model, and segmenting the inner ear structure into: external semicircular canal, posterior semicircular canal, superior semicircular canal, total canal, vestibule, cochlea and eyeball.
Based on the inner ear structure segmentation result, the method converts the center line of each semicircular canal into a point group to construct a space coordinate system, and specifically comprises the following steps:
s21, converting the inner ear central line model into a point array from the segmented temporal bone image, and obtaining the central ring intersection point formed by the central lines of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal. Specifically, VMTK (VMTK is an open source c++ library using ITK and VTK for segmenting, extracting and analyzing vascular structures, and has an insert for providing 3D slice software) is utilized to sequentially calculate inner tangent spheres of each semicircular canal, and the center of each inner tangent sphere forms the center line of the corresponding semicircular canal. According to analysis and research, the central line in the whole inner ear space comprises three central rings of semicircular tubes, and common crossing points are arranged among the central rings, so that the crossing points occur at least 3 times in an array, wherein the distance between the crossing points of the central rings of the semicircular tubes is provided with a fixed mode, and therefore other abnormal crossing points can be eliminated.
S22, calculating the connecting line distance of each point between the intersecting points, removing abnormal intersecting points according to the length of the central ring of each semicircular tube, and identifying the space positions of the outer semicircular tube, the rear semicircular tube and the upper semicircular tube, wherein the intersecting point above the rear semicircular tube and the upper semicircular tube is correspondingly the top end of the main pipe. In the inner ear space structure, the ring length of the center ring is sequentially from large to small and is a rear semicircular canal, an upper semicircular canal and an outer semicircular canal, the distance between the corresponding semicircular canal center ring cross points is provided with a fixed mode, accordingly, the space positions of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal corresponding to the connecting line distance of each point between the cross points can be identified, and through research on a large number of samples, the plane formed by the lowest edge of the bilateral eyeball and the top end of the common tube of the bilateral semicircular canal, namely, the fundus plane of the semicircular canal can be identified as a horizontal plane, so that the cross point above the rear semicircular canal and the upper semicircular canal in the point group corresponds to the top end of a main pipe. The top end of the total tube and the bottom of the eyeball are positioned in the cross section through rotating the segmented temporal bone image, and the bilateral inner ears are symmetrical relative to the sagittal plane.
s23. taking the lowest edge of the two-sided eyeball in the inner ear, namely the lowest point of the eyeball coordinates (the position with the minimum Z value) and the top ends of the two-sided main pipes to form a plane, calculating a plane equation of the plane, calculating and analyzing the distance from each point of the eyeball to the plane through the plane equation, taking the point with the largest distance, finally taking the point and the top ends of the two-sided main pipes to form a semicircular tube eyeball plane as a cross section, wherein the cross section represents the horizontal plane of a natural head-up position, and taking the normal vector of the cross section as the Z axis.
S24, taking the connecting lines of the top ends of the main pipes at two sides as an X axis, and taking the cross of the X axis and the Z axis as a Y axis.
Specifically, firstly, a plane formed BY the lowest edge of the bilateral eyeball and the top ends of the two-side main pipe is obtained BY adopting a least square fitting point array, for a discrete point array, a plane Z=AX+BY+C needs to be determined, so that the distance between the plane and each point in the point array is nearest, a matrix equation is constructed based on the object, the plane equation is solved to be AX+BY+CZ+D=0, and the normal vector of the plane is (A, B and C). The least squares method (also known as least squares) is a mathematical optimization technique. It finds the best functional match for the data by minimizing the sum of squares of the errors. According to the gaussian-Markov (Gauss-Markov) theorem, a common least squares estimator is best compared to any linear unbiased estimator found by other methods. The distance from the eyeball to the plane is further analyzed through the determined basic plane, so that a semicircular canal eyeball plane formed by the eyeball bottom and the top ends of the two side main pipes is more accurately determined, the segmented temporal bone image is guided to rotate to the horizontal plane of the natural head-up position, and the opposite sagittal planes of the two sides of the inner ear are symmetrical, and at the moment, the normal vector of the semicircular canal eyeball plane is taken as the Z axis. At this time, according to the symmetry of the left semicircular canal and the right semicircular canal in the inner ear structure, a coronal plane can be established, the top ends of the main pipes on two sides of the symmetry are continuously taken as X axes, and the cross multiplication of the X axes and the Z axes is taken as Y axes, so that the establishment of a standard space coordinate system is completed, through the space coordinate system, the inner ear space structure can be accurately positioned through coordinates, and the analysis of the inner ear space posture is facilitated, so that the semicircular canal posture mathematical model is constructed.
In a preferred embodiment, the method further comprises a construction method based on the individual semicircular canal space posture mathematical model after the step S3, and the human semicircular canal space posture mathematical model is obtained by detecting multiple groups of bilateral inner ear models through calculation, wherein the human semicircular canal space posture mathematical model is composed of average unit normal vectors of the planes of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal, and the average unit normal vectors are obtained through the following steps:
obtaining at least 100 divided bilateral inner ear models, obtaining the intersection of central rings formed by central lines of an outer semicircular canal, a rear semicircular canal and an upper semicircular canal in each model to obtain an inner ear space point array set, and adopting a least square method to fit planes of the semicircular canals respectively to obtain a plane equation. And (2) respectively calculating the direction angle of each semicircular canal plane in the bilateral inner ear model and each coordinate axis in the space coordinate system of the step (S2), and establishing a unit normal vector according to the cosine of the direction angle. And calculating the unit vector of the vector sum of the unit normal vectors of all semicircular canals of all at least 100 divided bilateral inner ear models, wherein the unit vector is the average unit normal vector of all semicircular canals, and the average unit normal vector of each semicircular canal plane obtained by the unit vector is used as a mathematical model of the space posture of the human body semicircular canals.
The human body semicircular canal space posture mathematical model constructed according to the space posture analysis method is positioned in a standard space coordinate system, and can directly generate a BPPV diagnosis and treatment scheme to guide the BPPV treatment operation.
Based on the spatial gesture analysis of the inner ear of the individual, the method for generating the personalized accurate BPPV diagnosis and treatment comprises the following steps: 1) individualizing the low head prone Dix-halpike test, 2) individualizing the 60 degree horizontal roll test, 3) individualizing the Epley reposition method, 4) individualizing the roll reposition method. The diagnosis and treatment test design principle of the BPPV comprises (1) the plane of the semicircular canal to be checked and the gravity direction are parallel (2) the top end of the ampulla ridge at the initial position is upward, which is beneficial to the entrance and exit of the otolith in the ampulla.
Taking an individualized low-head horizontal Dix-Hall pike test as an example, the traditional Dix-Hall pike test operation is that a patient is firstly in a sitting position, the head rotates to one side to enable the posterior semicircular canal on the same side to be parallel to the sagittal plane, then the head is laid down to lean backwards to form an angle of 20-30 degrees below the table surface of the examination table and the horizontal plane, and the otolith moves away from the ampulla under the action of gravity to generate excitatory stimulation.
In order to better determine the diagnosis and treatment method of the patient, temporal bone image data generated in the operation process are substituted into a human body semicircular canal space posture mathematical model constructed by the method, an individualized diagnosis and treatment method can be designed according to the measured semicircular canal plane normal vector, specifically, head rotation is carried out according to an Euler angle scheme from rotation to sagittal plane parallelism calculated according to the rear semicircular canal plane normal vector, cosine average values of included angles between the rear semicircular canal and coordinate axes of the human body semicircular canal are calculated according to the rear semicircular canal plane normal vector of the patient in the human body semicircular canal space posture mathematical model, and the Euler angle scheme from rotation of the rear semicircular canal to sagittal plane parallelism is obtained to guide the head rotation of the patient so that the rear semicircular canal on the same side is parallel to the sagittal plane, then the head is rotated by 60 degrees around the X axis to 60 degrees, the head is quickly laid down to the downward position of the right ear after sitting, and the head is not reclined.
And a path for rotating to the sagittal plane parallel can be calculated by smooth interpolation according to the normal vector of the posterior semicircular canal plane, and the machine reset can be used.
Taking an individualized 60-degree horizontal rolling test as an example, calculating an included angle cosine average value between an outer semicircular canal and each coordinate axis by using an outer semicircular canal plane normal vector of a patient in the human body semicircular canal space posture mathematical model to obtain an Euler angle scheme that the outer semicircular canal rotates to be parallel to a coronal plane, and guiding the patient to perform the following operations: A. the head is rotated to the outer semicircular canal and coronal plane parallel b. 60 degrees to one side (60 degrees to the Y axis) c. 120 degrees to the other side (120 degrees to the Y axis) d. 120 degrees to the other side (120 degrees to the Y axis).
The external semicircular canal plane normal vector or the average external semicircular canal normal vector of the patient is calculated through the human body semicircular canal space posture mathematical model, the external semicircular canal plane is accurately rotated to be parallel to the gravity direction, the operation method is effectively improved, and the following beneficial effects can be achieved: 1. the left and right turning head is convenient to operate by 60 degrees, the turning can be avoided, the operation is easy to complete, and the cervical vertebra injury can be avoided; 2. the amplitude of the turning head is 120 degrees, the movement path of the otolith is long, and the sensitivity is increased; 3. the contralateral long arm kerbstone can not be reset.
Based on the same concept as that of the above embodiments of the inner ear spatial posture analysis method of the present invention, the inner ear spatial posture analysis system provided by the embodiment of the present invention is described below, and the inner ear spatial posture analysis system described below and the inner ear spatial posture analysis method described above may be referred to correspondingly with each other. Referring to fig. 2, an embodiment of the present invention provides an inner ear spatial gesture analysis system, including: the system comprises a segmentation module, a space coordinate system construction module, a model construction module and a scheme making module, wherein an automatic segmentation model is constructed through the inner ear space posture analysis system so as to realize automatic segmentation of temporal bone image data, and an inner ear space standard three-dimensional coordinate system is constructed so as to realize accurate positioning of the inner ear space position, so that a human body semicircular space posture mathematical model can be obtained through intelligent training through a large number of sample collection, and the rapid diagnosis of a patient can be realized through the model and a personalized accurate BBPV diagnosis and treatment scheme can be directly generated.
The segmentation module is used for carrying out model segmentation on the structures of the outer semicircular canal, the rear semicircular canal, the upper semicircular canal, the main pipe, the vestibule, the cochlea and the eyeball in the temporal bone image; the space coordinate system construction module is used for establishing a horizontal plane to construct an inner ear space coordinate system by using the semicircular canal fundus plane in the segmented temporal bone image; the model construction module is used for calculating unit vectors of plane normal vectors of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal in a space coordinate system and constructing a human body semicircular canal space posture mathematical model; the scheme making module is used for calculating a human body semicircular canal space posture mathematical model, analyzing temporal bone image data of the BBPV patient and making an individualized BBPV diagnosis and treatment scheme of the patient.
The space coordinate system constructed by the space coordinate system construction module of the inner ear space attitude analysis system takes a plane normal vector formed by the lowest edge of the bilateral eyeball of the head of a human body and the top end of the bilateral semicircular canal common pipe as a Z axis, takes the top end connecting line of the bilateral semicircular canal common pipe as an X axis and takes the cross multiplication of the X axis and the Z axis as a Y axis, and can establish a semicircular canal space attitude mathematical model which can be directly applied and is positioned in the space coordinate system by the space coordinate system, so that the following problems can be solved: 1. automatic segmentation semicircular canal and eyeball based on depth operation; 2. taking a central line by an automatic semicircular tube; 3. automatically constructing a standard space coordinate system; 4. and establishing a semicircular canal posture mathematical model to automatically complete vestibular function examination and automatically generate a BPPV diagnosis and treatment scheme.
In a preferred embodiment, the spatial coordinate system construction module of the inner ear spatial pose analysis system of the present invention includes a conversion unit, configured to obtain a center line of each semicircular canal in the segmented temporal bone image, convert the inner ear center line model into a point array, and identify a spatial position by analyzing an intersection of center rings of each semicircular canal. The central line of the whole inner ear comprises three semicircular tube central rings, common crossing points are arranged among the rings, the crossing points appear at least 3 times in the array, wherein the distance between the crossing points of the semicircular tube central rings is provided with a fixed mode, so that other abnormal crossing points can be eliminated, and the space position of the inner ear can be accurately positioned.
The foregoing is merely a preferred embodiment of the present invention, and it should be noted that modifications and additions may be made to those skilled in the art without departing from the method of the present invention, which modifications and additions are also to be considered as within the scope of the present invention.
Claims (8)
1. An inner ear space gesture analysis method is characterized by comprising the following steps:
constructing a segmentation model according to temporal bone image data, wherein the temporal bone structure segmented by the model comprises: outer semicircular canal, posterior semicircular canal, superior semicircular canal, main canal, vestibule, cochlea and eyeball;
the method comprises the steps of rotating a segmented temporal bone image to enable the top end of a main pipe and the bottom of an eyeball to be located in a cross section, enabling two-side inner ears to be symmetrical relative to a sagittal plane, selecting a plane formed by the lowest edges of the two-side eyeballs and the top ends of two-side semicircular canals in the segmented temporal bone image, taking the plane as a horizontal plane of a space coordinate system, and enabling normal vectors of the plane to be a Z axis of the space coordinate system; taking the connecting lines at the top ends of the manifolds at two sides as X axes, taking the cross multiplication of the X and Z axes as Y axes, and constructing a space coordinate system according to the X and Z axes;
obtaining central lines of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal from the segmented temporal bone image, converting the central lines into point arrays, adopting a least square method to fit planes respectively to obtain plane equations of the semicircular canals, calculating unit vectors of plane normal vectors of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal in the space coordinate system, and constructing an individual semicircular canal space posture mathematical model;
detecting a plurality of double-sided inner ear models, calculating unit vectors of the sum of normal vectors of all semicircular canal planes of the double-sided inner ear models in the space coordinate system, using the unit vectors as a human body semicircular canal space posture mathematical model to analyze temporal bone image data of a BBPV patient, and making an individualized BBPV diagnosis and treatment scheme of the patient.
2. The method for analyzing the spatial pose of the inner ear according to claim 1, wherein the step of constructing the segmentation model according to temporal bone image data is realized by:
deriving temporal bone image sample data from a medical image information system, and storing the temporal bone image sample data in a Dicom format;
performing recognition training on an inner ear structure and an eyeball in temporal bone image sample data by adopting a 3D convolutional neural network, and constructing an automatic segmentation model of a semicircular canal and the eyeball;
the temporal bone image is automatically segmented through the automatic segmentation model, and the inner ear structure is segmented into: external semicircular canal, posterior semicircular canal, superior semicircular canal, total canal, vestibule, cochlea and eyeball.
3. The inner ear spatial pose analysis method according to claim 1, wherein the construction process of the Z-axis of the spatial coordinate system comprises:
respectively acquiring central lines of each outer semicircular canal, each rear semicircular canal and each upper semicircular canal in the inner ear from the segmented temporal bone image;
acquiring the intersection point of a central ring formed by the central lines of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal, and converting the inner ear central line model into a point array;
calculating the connection distance of each point between the intersecting points, removing abnormal intersecting points according to the length of the central ring of each semicircular tube, and identifying the spatial positions of the outer semicircular tube, the rear semicircular tube and the upper semicircular tube, wherein the intersecting point above the rear semicircular tube and the upper semicircular tube is correspondingly the top end of a main pipe;
taking the lowest point of eyeball coordinates in the inner ear and the top ends of the main pipes on two sides to form a plane and calculating a plane equation;
and analyzing the distance between each point of the eyeball and the plane, and taking the point with the largest distance, wherein the point and the top ends of the main pipes on two sides form a semicircular canal eyeball plane as a cross section, and the normal vector of the cross section is a Z axis.
4. The inner ear spatial pose analysis method according to claim 1, further comprising a construction method based on the individual semicircular canal spatial pose mathematical model, wherein a human body semicircular canal spatial pose mathematical model is obtained by detecting a plurality of sets of bilateral inner ear models, the human body semicircular canal spatial pose mathematical model is composed of average unit normal vectors of planes of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal, and the average unit normal vector is obtained by:
obtaining at least 100 divided double-sided inner ear models, respectively calculating the direction angle of each semicircular tube plane in the double-sided inner ear models and each coordinate axis in the space coordinate system, and establishing a unit normal vector according to the cosine of the direction angle;
and calculating a unit vector of the vector sum of the unit normal vectors, wherein the unit vector is an average unit normal vector.
5. The inner ear spatial pose analysis method according to claim 4, characterized in that said inner ear spatial pose analysis method further comprises: and analyzing temporal bone image data of the BBPV patient according to the human body semicircular canal space posture mathematical model, and making an individualized BBPV diagnosis and treatment scheme of the patient.
6. The method of claim 5, wherein the personalized BBPV diagnostic protocol comprises:
calculating cosine average values of included angles between the rear semicircular canal and each coordinate axis by using the plane normal vector of the rear semicircular canal of the patient in the human body semicircular canal space posture mathematical model to obtain Euler angle schemes and rotation paths from the rotation of the rear semicircular canal to the sagittal plane parallelism, and/or
And calculating cosine average values of included angles between the outer semicircular canal and each coordinate axis by using the outer semicircular canal plane normal vector of the patient in the human body semicircular canal space posture mathematical model, and obtaining an Euler angle scheme and a rotating path of the outer semicircular canal rotating to the coronal plane parallel.
7. An inner ear spatial pose analysis system, comprising:
the segmentation module is used for carrying out model segmentation on the structures of the outer semicircular canal, the rear semicircular canal, the upper semicircular canal, the main pipe, the vestibule, the cochlea and the eyeball in the temporal bone image;
the space coordinate system construction module is used for constructing an inner ear space coordinate system by taking a plane normal vector formed by the lowest edge of the bilateral eyeballs of the head of a human body and the top end of the bilateral semicircular canal common pipe as a Z axis, taking a top end connecting line of the bilateral semicircular canal common pipe as an X axis and taking the cross multiplication of the X axis and the Z axis as a Y axis;
the model construction module is used for calculating unit vectors of plane normal vectors of the outer semicircular canal, the rear semicircular canal and the upper semicircular canal in a space coordinate system and constructing a mathematical model of the space posture of the individual semicircular canal;
the scheme making module is used for calculating a human body semicircular canal space posture mathematical model, analyzing temporal bone image data of the BBPV patient and making an individualized BBPV diagnosis and treatment scheme of the patient.
8. The inner ear spatial pose analysis system according to claim 7, wherein the spatial coordinate system construction module comprises a conversion unit for acquiring a center line of each semicircular canal in the segmented temporal bone image, converting the inner ear center line model into a point array, and identifying the spatial position by analyzing the intersection point of each semicircular canal center ring.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010935021.8A CN112288687B (en) | 2020-09-08 | 2020-09-08 | Inner ear space posture analysis method and analysis system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010935021.8A CN112288687B (en) | 2020-09-08 | 2020-09-08 | Inner ear space posture analysis method and analysis system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112288687A CN112288687A (en) | 2021-01-29 |
| CN112288687B true CN112288687B (en) | 2024-04-09 |
Family
ID=74420344
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010935021.8A Active CN112288687B (en) | 2020-09-08 | 2020-09-08 | Inner ear space posture analysis method and analysis system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112288687B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114187296B (en) * | 2021-11-09 | 2022-12-13 | 元化智能科技(深圳)有限公司 | Capsule endoscope image focus segmentation method, server and system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6609523B1 (en) * | 1999-10-26 | 2003-08-26 | Philip F. Anthony | Computer based business model for a statistical method for the diagnosis and treatment of BPPV |
| CN103932711A (en) * | 2014-04-23 | 2014-07-23 | 田永胜 | Inner-ear posture adjustment monitoring system based on virtual reality and inner-ear posture capturing device |
| CN107423773A (en) * | 2016-05-23 | 2017-12-01 | 北京师范大学 | The autoegistration method and device of three-dimensional cranium |
| CN107945868A (en) * | 2017-11-24 | 2018-04-20 | 中国科学院苏州生物医学工程技术研究所 | Benign paroxysmal positional vertigo intelligence diagnostic equipment |
| CN108364347A (en) * | 2018-03-16 | 2018-08-03 | 温州市人民医院 | A kind of labyrinthus membranaceus model production method |
| CN108926426A (en) * | 2018-07-11 | 2018-12-04 | 温州市人民医院 | A kind of BBPV real-time three-dimensional navigation system |
| CN110544264A (en) * | 2019-08-28 | 2019-12-06 | 北京工业大学 | Temporal bone key anatomical structure small target segmentation method based on 3D deep supervision mechanism |
| CN110969698A (en) * | 2019-12-23 | 2020-04-07 | 首都医科大学附属北京友谊医院 | Construction method of temporal bone space coordinate system, space positioning method and electronic equipment |
| CN111354073A (en) * | 2019-12-23 | 2020-06-30 | 首都医科大学附属北京友谊医院 | Construction method, spatial positioning method and equipment of unilateral temporal bone spatial coordinate system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9937092B2 (en) * | 2013-08-06 | 2018-04-10 | Sau Lan Tang Staats | Device and methods for positioning the head of a reclining person |
-
2020
- 2020-09-08 CN CN202010935021.8A patent/CN112288687B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6609523B1 (en) * | 1999-10-26 | 2003-08-26 | Philip F. Anthony | Computer based business model for a statistical method for the diagnosis and treatment of BPPV |
| CN103932711A (en) * | 2014-04-23 | 2014-07-23 | 田永胜 | Inner-ear posture adjustment monitoring system based on virtual reality and inner-ear posture capturing device |
| CN107423773A (en) * | 2016-05-23 | 2017-12-01 | 北京师范大学 | The autoegistration method and device of three-dimensional cranium |
| CN107945868A (en) * | 2017-11-24 | 2018-04-20 | 中国科学院苏州生物医学工程技术研究所 | Benign paroxysmal positional vertigo intelligence diagnostic equipment |
| CN108364347A (en) * | 2018-03-16 | 2018-08-03 | 温州市人民医院 | A kind of labyrinthus membranaceus model production method |
| CN108926426A (en) * | 2018-07-11 | 2018-12-04 | 温州市人民医院 | A kind of BBPV real-time three-dimensional navigation system |
| CN110544264A (en) * | 2019-08-28 | 2019-12-06 | 北京工业大学 | Temporal bone key anatomical structure small target segmentation method based on 3D deep supervision mechanism |
| CN110969698A (en) * | 2019-12-23 | 2020-04-07 | 首都医科大学附属北京友谊医院 | Construction method of temporal bone space coordinate system, space positioning method and electronic equipment |
| CN111354073A (en) * | 2019-12-23 | 2020-06-30 | 首都医科大学附属北京友谊医院 | Construction method, spatial positioning method and equipment of unilateral temporal bone spatial coordinate system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112288687A (en) | 2021-01-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111728609B (en) | Electroencephalogram signal classification method, classification model training method, device and medium | |
| WO2021078064A1 (en) | Ultrasonic scanning track planning method and apparatus, and storage medium and computer device | |
| Sforza et al. | Soft-tissue facial anthropometry in three dimensions: from anatomical landmarks to digital morphology in research, clinics and forensic anthropology | |
| CN112509119B (en) | Spatial data processing and positioning method and device for temporal bone and electronic equipment | |
| JP2008534082A (en) | Systems, methods and media for acquiring and generating standard ultrasound images independent of operators of fetal, neonatal and adult organs | |
| CN112308964B (en) | Spatial data processing and positioning method, device and equipment for unilateral temporal bone | |
| CN101692286A (en) | Method for acquiring three-view drawing of medical image | |
| CN112001122A (en) | Non-contact physiological signal measuring method based on end-to-end generation countermeasure network | |
| CN112288687B (en) | Inner ear space posture analysis method and analysis system | |
| Tian et al. | Efficient tooth gingival margin line reconstruction via adversarial learning | |
| Pepe et al. | Pattern recognition and mixed reality for computer-aided maxillofacial surgery and oncological assessment | |
| Gao et al. | 3DSRNet: 3D Spine Reconstruction Network Using 2D Orthogonal X-ray Images Based on Deep Learning | |
| CN111445575A (en) | Image reconstruction method and device of Wirisi ring, electronic device and storage medium | |
| Smith et al. | Automated measurement of pediatric cranial bone thickness and density from clinical computed tomography | |
| CN117649442B (en) | Method for constructing three-dimensional head shadow measurement coordinate system by utilizing anatomical landmarks of balance organ | |
| KR101309013B1 (en) | Diagnosis Device For Face Form Using Facial Image and Cephalometric Image | |
| CN115249290B (en) | Spatial data processing method, spatial positioning method and device for unilateral temporal bone | |
| CN107170009B (en) | Medical image-based goggle base curve data measurement method | |
| KR100960893B1 (en) | Method for comparing 3D models to classify Sasang constitution | |
| US20220338836A1 (en) | System and method for guiding positioning and orienting of an ultrasound probe | |
| Yang et al. | Fully automatic spinal scanning and measurement based on multi-source vision information | |
| CN118351110B (en) | Chloasma severity objectively evaluating method based on multitask learning | |
| CN118247219B (en) | Tooth root external absorption detection system based on deep learning | |
| Wu et al. | Mathematical Model of Human Semicircular Canal Spatial Attitude | |
| US20240293103A1 (en) | System and method for guiding positioning and orienting of an ultrasound probe |
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 |