Nothing Special   »   [go: up one dir, main page]

CN113538464A - Brain image segmentation model training method, segmentation method and device - Google Patents

Brain image segmentation model training method, segmentation method and device Download PDF

Info

Publication number
CN113538464A
CN113538464A CN202110832623.5A CN202110832623A CN113538464A CN 113538464 A CN113538464 A CN 113538464A CN 202110832623 A CN202110832623 A CN 202110832623A CN 113538464 A CN113538464 A CN 113538464A
Authority
CN
China
Prior art keywords
brain
patient
segmentation model
image
training
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.)
Pending
Application number
CN202110832623.5A
Other languages
Chinese (zh)
Inventor
隋轶
刘婷
李洪浩
徐冰
张忠
杨鸣
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Naoxi Suzhou Intelligent Technology Co ltd
Original Assignee
Naoxi Suzhou Intelligent Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Naoxi Suzhou Intelligent Technology Co ltd filed Critical Naoxi Suzhou Intelligent Technology Co ltd
Priority to CN202110832623.5A priority Critical patent/CN113538464A/en
Publication of CN113538464A publication Critical patent/CN113538464A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • 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
    • 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
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • 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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G06T2207/10081—Computed x-ray tomography [CT]
    • 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
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G06T2207/10088—Magnetic resonance imaging [MRI]
    • 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
    • G06T2207/30016—Brain

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Public Health (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Medical Informatics (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Primary Health Care (AREA)
  • Molecular Biology (AREA)
  • Epidemiology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Pathology (AREA)
  • Databases & Information Systems (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

The invention provides a brain image segmentation model training method, a segmentation method and a device, wherein the method comprises the following steps: acquiring a scanning image of a brain of a patient and a brain partition template corresponding to symptoms of the patient; establishing a neural network based on the scanning image and the data set of the brain partition template, and training a neural network segmentation model; and weighting the brain subareas in the brain subarea template to obtain a brain image segmentation model. The invention takes the symptoms of the apoplexy patient to correspond to the diseased brain area, and takes the brain area as high weight to be combined with the image of the apoplexy patient during the treatment for carrying out segmentation and labeling. The exact brain region corresponding to the symptom can be found.

Description

Brain image segmentation model training method, segmentation method and device
Technical Field
The invention relates to the technical field of image processing, in particular to a brain image segmentation model training method, a brain image segmentation method and a brain image segmentation model training device.
Background
Acute or chronic occlusion of cerebral arteries results in infarction with insufficient blood supply to a part of the brain tissue, resulting in ischemic stroke. Based on medical images such as CT or magnetic resonance, segmentation and labeling are carried out on the infarct area of the brain tissue, and the segmentation and labeling are a hot spot in medical image processing. However, ischemic stroke is likely to recur, and imaging of old infarcts and new infarct areas appears very close to CT flat-scan, mr diffusion-weighted images. The existing image processing method is difficult to distinguish.
Disclosure of Invention
In view of the above, the present invention provides a method, a device and a system for training a brain image segmentation model, so as to solve the technical problems in the background art.
In a first aspect, the present invention provides a method for training a brain image segmentation model, including: acquiring a scanning image of a brain of a patient and a brain partition template corresponding to symptoms of the patient;
establishing a neural network based on the scanning image and the data set of the brain partition template, and training a neural network segmentation model;
and weighting the brain subareas in the brain subarea template to obtain a brain image segmentation model.
According to one embodiment of the invention, an attention mechanism is added to the decoding layer of the neural network segmentation model.
According to one embodiment of the present invention, obtaining a brain partition template corresponding to a symptom of a patient comprises:
obtaining a symptom of the patient;
based on the patient's symptoms, a brain partition template corresponding to the symptoms is found in a set of pre-set brain partitions.
According to one embodiment of the invention, the patient's symptoms include one or more of visual paralysis, expressive aphasia, hemineglect, perceptual aphasia, disorientation, central or hypoglossal nerve paralysis, ipsilateral hemianopsia, upper limb paralysis, lower limb paralysis, upper limb sensory loss, lower limb sensory loss, hemifacial sensory loss, and dysarthria.
According to one embodiment of the invention, the brain partition comprises: thalamus, corona radiata, basal ganglia, anterior cerebral artery supply, frontal pole to central anterior sulcus region, anterior cerebral artery supply, lateral central leaflet region, anterior cerebral artery supply, anterior neocortical wedge and apical leaflet region, middle cerebral artery supply partition and posterior cerebral artery supply partition, wherein one said symptom correspondence includes at least one brain partition.
According to an embodiment of the invention, the scan images comprise one or more of CT scout, CT enhancement, CT perfusion, MR diffusion, MR perfusion and MR-spin echo labeling.
In a second aspect, the present invention further provides a device for training a cerebral infarction image segmentation model, including:
the acquisition module is used for acquiring a scanning image of the brain of the patient and a brain subarea template corresponding to symptoms of the patient;
the training module is used for establishing a neural network based on the scanning image and the data set of the brain partition template and training a neural network segmentation model;
and the determining module is used for weighting the brain subareas in the brain subarea template to obtain a brain infarction image segmentation model.
According to one embodiment of the invention, the determining module is configured to: and adding an attention mechanism into a decoding layer of the neural network segmentation model.
According to an embodiment of the present invention, the obtaining module is configured to:
obtaining a symptom of the patient;
based on the patient's symptoms, a brain partition template corresponding to the symptoms is found in a set of pre-set brain partitions.
In a third aspect, the present invention further provides a method for segmenting a brain image, including:
acquiring a scanning image to be processed;
the scan image is segmented by a brain image segmentation model, wherein the brain image segmentation model is obtained by the method of the embodiment of the first aspect.
In a fourth aspect, the present invention further provides a brain image segmentation apparatus, including:
the acquisition module is used for acquiring a to-be-processed scanning image;
and the processing module is used for carrying out segmentation processing on the scanning image through a brain image segmentation model, wherein the brain image segmentation model is obtained by training the method in the embodiment of the first aspect.
In a fifth aspect, the present invention provides an electronic device, comprising:
a memory for storing instructions for execution by one or more processors of the device, an
A processor configured to perform the method of the first aspect of the above embodiments.
In a sixth aspect, the present invention provides a computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, causes the processor to execute the method of the first aspect.
Drawings
FIG. 1 is a flow chart of a method for training a brain image segmentation model according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a brain partition template based on a standard brain according to an embodiment of the present invention;
FIG. 3 is a structural feature diagram of a dual-input deep neural network based on U-NET and attention mechanism according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
The brain image segmentation model training method provided by the embodiment of the invention can be applied to electronic equipment. Such as medical test equipment.
The ischemic stroke is easy to recur, and the imaging expression of old infarction and new infarct area is very close to that of CT flat scan and magnetic resonance diffusion weight images, so that the two images are difficult to distinguish by the prior art, and the false positive is difficult to accurately judge. The invention takes the symptoms of the apoplexy patient to correspond to the diseased brain area, and takes the brain area as high weight to be combined with the image of the apoplexy patient during the treatment for carrying out segmentation and labeling. The problem of false positive based on image only in the prior art can be effectively reduced.
Fig. 1 shows a flowchart of the brain image segmentation model training method of the present invention, which includes the following steps:
s110, a scanning image of the brain of the patient and a brain subarea template corresponding to the symptoms of the patient are obtained.
Taking the cerebral infarction as an example, in the process of developing medical care, a large number of scanned images of the brain related to the cerebral infarction are stored, and the scanned images are used as first input data of model training. For example, the scan image may be an image sensitive to cerebral infarction imaging, such as CT flat scan, CT enhancement, CT perfusion, MR diffusion, MR perfusion, MR-spin echo labeling, and the like.
The brain subarea template corresponding to the symptoms of the patient corresponds to the subarea according to the clinical performance. As shown in table 1.
TABLE 1 correspondence table of symptoms and infarcted brain area
Brain partition Symptoms and signs
M1,M4 Staring or paralytic vision
M1,M4 Expressive aphasia
A3,M6 Half side neglect
M3,M6 Loss of consciousness
A1,M1,M4 Disorientation
M5,CR,BG Paralysis of central facial nerve or hypoglossal nerve
P,M2,M3 Ipsilateral hemianopsia
M5,CR,BG Paralysis of upper limbs
A2,CR,BG Paralysis of lower limbs
M5,CR,BG,T Sensory loss of the upper limbs
A2,CR,BG,T Sensory loss of lower limbs
M5,CR,T Loss of semi-lateral facial sensation
A1,M1,M4,BG Dysarthria
Wherein: t is Thalamus Thalamus, CR is corona radiata, BG is BasalGanglion basal ganglia, A1 is anterior cerebral artery blood supply area, frontal pole to central anterior sulcus area, A2 is anterior cerebral artery blood supply area, central paraleaflet area, A3 is anterior cerebral artery blood supply area, anterior neocortical wedge and apical leaflet area, M1-M6 is the classic middle cerebral artery blood supply area partition (refer to ASPECTS scoring area partition), P is posterior cerebral artery blood supply area, including occipital pole, medial cerebral cortex and medial basal part of temporal cerebral lobe.
FIG. 2 is a schematic diagram of a brain partition template based on a standard brain according to the present invention. It can be seen from fig. 2 that the brain is divided into a plurality of regions, i.e. the regions correspond to the divisions of table 1, and the symptoms in table 1 can correspond to the occurrence of occlusions in the divisions, and the corresponding brain region correspondence is obtained according to the symptoms of the patient.
And S120, establishing a neural network based on the scanning image and the data set of the brain partition template, and training a neural network segmentation model.
In one embodiment of the present invention, the neural network may be a deep neural network or the like. Preferably, it may be a U-NET neural network.
Referring to fig. 3, fig. 3 shows a structural feature diagram of a dual-input deep neural network based on a U-NET and an attention mechanism. As shown in fig. 3, a scanned image of a stroke patient is input 1 (first input data), a brain template corresponding to a symptom is input 2 (second input data), 5 × 5 convolution processing is performed on the data of the input 1, 3 × 3 convolution processing is performed on the data of the input data 2 for feature extraction, extracted features are fused, and a result is output. For example, the infarct area segmentation labeling.
Wherein, the coding block in fig. 3 is defined as:
Figure BDA0003176081440000051
where f is a convolution operation, expressed as:
Figure BDA0003176081440000052
g is the pooling operation, expressed as:
Figure BDA0003176081440000053
where X is the input, K is the convolution kernel, and 5 × 5 coded blocks, i.e., K is a convolution kernel of size 5 × 5.
The decoding block is defined as:
Figure BDA0003176081440000054
where phi is defined as the upsampling operation of the bilinear interpolation.
The MAM module is defined as:
Figure BDA0003176081440000055
αX＝σ(qatt(X,G；θatt))
Figure BDA0003176081440000056
let the input image be I, the template drawing be A, the layer d where the coding block is located,
the inputs of the decoding block of the ith layer of the template graph branch are:
EA,d-i(A)+φ(DA，i-1(A))
wherein D is.,i-1And E.,dSame, E.,0I.e. feature maps of the same size as the input template map matrix.
The inputs to the image branch (upper half) layer i decoding block are:
EI,i-i(I)+φ(DI,i-1(I),DA,i-1(A))
the whole network is represented as Φ, and the final output of the network is:
Φ(I,A)＝f(DI,d(I)×LBS(DA,d(A)))
wherein LBS stands for using step linear normalization and sigmoid activation layers.
And S130, weighting the brain subareas in the brain subarea template to obtain a brain image segmentation model. For example: with reference to table 1, if the patient has paralyzed right upper limb, central facial nerve and lingual nerve, the M5, CR and BG area images are labeled as 1, and the rest are 0.
According to one embodiment of the invention, an attention mechanism is added to the decoding layer of the neural network segmentation model. Based on the mechanism, the region of interest extracted by the region detection network is used as attention information, so that the network concentrates the possible lesion region, the interference of other information can be avoided, the model accuracy can be improved, and the diagnosis efficiency can be improved.
In one embodiment of the present invention, obtaining a brain partition template corresponding to a symptom of a patient comprises: the symptoms of the patient are obtained, wherein the symptoms can be input manually or detected automatically by a machine. Based on the patient's symptoms, a brain partition template corresponding to the symptoms is found in a set of pre-set brain partitions.
According to the model obtained by the model training method provided by the embodiment of the invention, the symptoms of the stroke patient correspond to the diseased brain area, and the brain area is used as a high weight to be segmented and labeled in combination with the image of the stroke patient during the treatment. The problem of false positive based on image only in the prior art can be effectively reduced. For example, subarachnoid cyst and stroke infarction focus are represented as low-density shadow on CT flat scan, and other low-density shadows are mistaken for infarction by directly segmenting with the image. According to the invention, the cerebral area weight is carried out by combining symptoms, so that an accurate infarct focus corresponding to the symptoms can be found.
The invention also discloses a device for training the cerebral infarction image segmentation model, which comprises:
the acquisition module is used for acquiring a scanning image of the brain of the patient and a brain subarea template corresponding to symptoms of the patient;
the training module is used for establishing a neural network based on the scanning image and the data set of the brain partition template and training a neural network segmentation model;
and the determining module is used for weighting the brain subareas in the brain subarea template to obtain a brain infarction image segmentation model.
In one embodiment of the present invention, the determining module is configured to: and adding an attention mechanism into a decoding layer of the neural network segmentation model.
In an embodiment of the present invention, the obtaining module is configured to:
obtaining a symptom of the patient;
based on the patient's symptoms, a brain partition template corresponding to the symptoms is found in a set of pre-set brain partitions.
The functions of the modules in the apparatus of the present invention have been described in detail in the above embodiments, and refer to the method described in fig. 1, which is not described herein again.
In addition, the invention also discloses a brain image segmentation method, which comprises the following steps: acquiring a scanning image to be processed;
the scan image is segmented by a brain image segmentation model, wherein the brain image segmentation model is obtained by the method shown in fig. 1.
Referring to fig. 4, the present invention also provides an electronic device including:
a memory 410 for storing instructions for execution by one or more processors of the device, an
A processor 420 for performing the method explained in the above embodiment with reference to fig. 1.
The invention also provides a computer-readable storage medium, in which a computer program is stored, which, when executed by a processor, causes the processor to carry out the method explained in fig. 1 in the above-mentioned embodiment.
Embodiments of the disclosed mechanisms may be implemented in hardware, software, firmware, or a combination of these implementations. Embodiments of the invention may be implemented as computer programs or program code executing on programmable systems comprising at least one processor, a storage system (including volatile and non-volatile memory and/or storage elements), at least one input device, and at least one output device.
Program code may be applied to input instructions to perform the functions described in this disclosure and to generate output information. The output information may be applied to one or more output devices in a known manner. For purposes of the present invention, a processing system includes any system having a Processor such as, for example, a Digital Signal Processor (DSP), a microcontroller, an Application Specific Integrated Circuit (ASIC), or a microprocessor.
The program code may be implemented in a high level procedural or object oriented programming language to communicate with a processing system. The program code can also be implemented in assembly or machine language, if desired. Indeed, the mechanisms described in this disclosure are not limited in scope to any particular programming language. In any case, the language may be a compiled or interpreted language.
In some cases, the disclosed embodiments may be implemented in hardware, firmware, software, or any combination thereof. The disclosed embodiments may also be implemented as instructions carried by or stored on one or more transitory or non-transitory machine-readable (e.g., computer-readable) storage media, which may be read and executed by one or more processors. For example, the instructions may be distributed via a network or via other computer readable media. Thus, a machine-readable medium may include any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computer), including, but not limited to, floppy diskettes, optical disks, Compact disk Read Only memories (CD-ROMs), magneto-optical disks, Read Only Memories (ROMs), Random Access Memories (RAMs), Erasable Programmable Read Only Memories (EPROMs), Electrically Erasable Programmable Read Only Memories (EEPROMs), magnetic or optical cards, flash Memory, or a tangible machine-readable Memory for transmitting information (e.g., carrier waves, infrared signals, digital signals, etc.) using the Internet in electrical, optical, acoustical or other forms of propagated signals. Thus, a machine-readable medium includes any type of machine-readable medium suitable for storing or transmitting electronic instructions or information in a form readable by a machine (e.g., a computer).
In the drawings, some features of the structures or methods may be shown in a particular arrangement and/or order. However, it is to be understood that such specific arrangement and/or ordering may not be required. Rather, in some embodiments, the features may be arranged in a manner and/or order different from that shown in the figures. In addition, the inclusion of a structural or methodical feature in a particular figure is not meant to imply that such feature is required in all embodiments, and in some embodiments, may not be included or may be combined with other features.
While the invention has been shown and described with reference to certain preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention.

Claims (10)

1. A brain image segmentation model training method is characterized by comprising the following steps:
acquiring a scanning image of a brain of a patient and a brain partition template corresponding to symptoms of the patient;
establishing a neural network based on the scanning image and the data set of the brain partition template, and training a neural network segmentation model;
and weighting the brain subareas in the brain subarea template to obtain a brain image segmentation model.
2. The training method of claim 1, wherein an attention mechanism is added to a decoding layer of the neural network segmentation model.
3. The training method of claim 1, wherein obtaining a brain partition template corresponding to a symptom of the patient comprises:
obtaining a symptom of the patient;
based on the patient's symptoms, a brain partition template corresponding to the symptoms is found in a set of pre-set brain partitions.
4. Training method according to any of claims 1-3, wherein the patient's symptoms comprise one or more of visual paralysis, expressive aphasia, hemineglect, perceptual aphasia, disorientation, central or hypoglossal nerve paralysis, ipsilateral hemianopsia, upper limb paralysis, lower limb paralysis, upper limb sensory loss, lower limb sensory loss, hemifacial sensory loss and dysarthria.
5. Training method according to claim 4, wherein the brain partition comprises: thalamus, corona radiata, basal ganglia, anterior cerebral artery supply, frontal pole to central anterior sulcus region, anterior cerebral artery supply, lateral central leaflet region, anterior cerebral artery supply, anterior neocortical wedge and apical leaflet region, middle cerebral artery supply partition and posterior cerebral artery supply partition, wherein one said symptom correspondence includes at least one brain partition.
6. The training method of claim 1, wherein the scan images comprise one or more of CT scout, CT enhancement, CT perfusion, MR diffusion, MR perfusion, and MR-spin echo labeling.
7. A brain infarction image segmentation model training device, includes:
the acquisition module is used for acquiring a scanning image of the brain of the patient and a brain subarea template corresponding to symptoms of the patient;
the training module is used for establishing a neural network based on the scanning image and the data set of the brain partition template and training a neural network segmentation model;
and the determining module is used for weighting the brain subareas in the brain subarea template to obtain a brain infarction image segmentation model.
8. The training apparatus of claim 7, wherein the determination module is configured to: and adding an attention mechanism into a decoding layer of the neural network segmentation model.
9. The training apparatus of claim 7, wherein the acquisition module is configured to:
obtaining a symptom of the patient;
based on the patient's symptoms, a brain partition template corresponding to the symptoms is found in a set of pre-set brain partitions.
10. A method for segmenting brain images, comprising:
acquiring a scanning image to be processed;
performing a segmentation process on the scan image by using a brain image segmentation model, wherein the brain image segmentation model is obtained by training according to the method of any one of claims 1 to 6.
CN202110832623.5A 2021-07-22 2021-07-22 Brain image segmentation model training method, segmentation method and device Pending CN113538464A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110832623.5A CN113538464A (en) 2021-07-22 2021-07-22 Brain image segmentation model training method, segmentation method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110832623.5A CN113538464A (en) 2021-07-22 2021-07-22 Brain image segmentation model training method, segmentation method and device

Publications (1)

Publication Number Publication Date
CN113538464A true CN113538464A (en) 2021-10-22

Family

ID=78088718

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110832623.5A Pending CN113538464A (en) 2021-07-22 2021-07-22 Brain image segmentation model training method, segmentation method and device

Country Status (1)

Country Link
CN (1) CN113538464A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114334130A (en) * 2021-12-25 2022-04-12 浙江大学 PET molecular image computer-aided diagnosis system based on brain symmetry
CN115908457A (en) * 2023-01-06 2023-04-04 脑玺(苏州)智能科技有限公司 Low-density peduncle area segmentation method, analysis method, device, system, equipment and medium

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102088910A (en) * 2008-07-07 2011-06-08 浜松光子学株式会社 Brain disease diagnosis system
CN109671069A (en) * 2018-12-14 2019-04-23 强联智创（北京）科技有限公司 A kind of measurement method and system of the core Infarction volume based on head CT images
CN111640095A (en) * 2020-05-22 2020-09-08 上海联影智能医疗科技有限公司 Method for quantifying cerebral microhemorrhage and computer readable storage medium
CN111667458A (en) * 2020-04-30 2020-09-15 杭州深睿博联科技有限公司 Method and device for detecting early acute cerebral infarction in flat-scan CT
CN111739647A (en) * 2020-08-07 2020-10-02 暨南大学 Magnetic resonance image-based bipolar disorder suicide risk prediction method and related device
CN112075927A (en) * 2020-10-15 2020-12-15 首都医科大学附属北京天坛医院 Method and device for classifying causes of cerebral apoplexy
CN112348785A (en) * 2020-10-29 2021-02-09 深圳大学 Epileptic focus positioning method and system
CN112508902A (en) * 2020-11-30 2021-03-16 上海联影智能医疗科技有限公司 White matter high signal grading method, electronic device and storage medium
CN112651976A (en) * 2020-10-13 2021-04-13 首都医科大学附属北京天坛医院 Focal region brain network determination method and system based on low-resolution nuclear magnetic data
CN112767403A (en) * 2021-01-08 2021-05-07 广东中科天机医疗装备有限公司 Medical image segmentation model training method, medical image segmentation method and device
CN112927240A (en) * 2021-03-08 2021-06-08 重庆邮电大学 CT image segmentation method based on improved AU-Net network
CN113077479A (en) * 2021-03-25 2021-07-06 深圳市铱硙医疗科技有限公司 Automatic segmentation method, system, terminal and medium for acute ischemic stroke focus

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102088910A (en) * 2008-07-07 2011-06-08 浜松光子学株式会社 Brain disease diagnosis system
CN109671069A (en) * 2018-12-14 2019-04-23 强联智创（北京）科技有限公司 A kind of measurement method and system of the core Infarction volume based on head CT images
CN111667458A (en) * 2020-04-30 2020-09-15 杭州深睿博联科技有限公司 Method and device for detecting early acute cerebral infarction in flat-scan CT
CN111640095A (en) * 2020-05-22 2020-09-08 上海联影智能医疗科技有限公司 Method for quantifying cerebral microhemorrhage and computer readable storage medium
CN111739647A (en) * 2020-08-07 2020-10-02 暨南大学 Magnetic resonance image-based bipolar disorder suicide risk prediction method and related device
CN112651976A (en) * 2020-10-13 2021-04-13 首都医科大学附属北京天坛医院 Focal region brain network determination method and system based on low-resolution nuclear magnetic data
CN112075927A (en) * 2020-10-15 2020-12-15 首都医科大学附属北京天坛医院 Method and device for classifying causes of cerebral apoplexy
CN112348785A (en) * 2020-10-29 2021-02-09 深圳大学 Epileptic focus positioning method and system
CN112508902A (en) * 2020-11-30 2021-03-16 上海联影智能医疗科技有限公司 White matter high signal grading method, electronic device and storage medium
CN112767403A (en) * 2021-01-08 2021-05-07 广东中科天机医疗装备有限公司 Medical image segmentation model training method, medical image segmentation method and device
CN112927240A (en) * 2021-03-08 2021-06-08 重庆邮电大学 CT image segmentation method based on improved AU-Net network
CN113077479A (en) * 2021-03-25 2021-07-06 深圳市铱硙医疗科技有限公司 Automatic segmentation method, system, terminal and medium for acute ischemic stroke focus

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114334130A (en) * 2021-12-25 2022-04-12 浙江大学 PET molecular image computer-aided diagnosis system based on brain symmetry
CN114334130B (en) * 2021-12-25 2023-08-22 浙江大学 Brain symmetry-based PET molecular image computer-aided diagnosis system
CN115908457A (en) * 2023-01-06 2023-04-04 脑玺(苏州)智能科技有限公司 Low-density peduncle area segmentation method, analysis method, device, system, equipment and medium

Similar Documents

Publication Publication Date Title
CN109993726B (en) Medical image detection method, device, equipment and storage medium
JP2007172604A (en) Method and apparatus for selecting computer-assisted algorithm based on protocol and/or parameter of acquisition system
CN113538464A (en) Brain image segmentation model training method, segmentation method and device
CN111145160A (en) Method, device, server and medium for determining coronary artery branch where calcified area is located
JP2023175011A (en) Document creation assistance device, method, and program
Che et al. Image quality-aware diagnosis via meta-knowledge co-embedding
JPH0773291A (en) Image processor
CN116309647B (en) Method for constructing craniocerebral lesion image segmentation model, image segmentation method and device
JP2019149094A (en) Diagnostic assistance system, diagnostic assistance method, and program
Sohail et al. A modified U-net based framework for automated segmentation of Hippocampus region in brain MRI
CN112767403A (en) Medical image segmentation model training method, medical image segmentation method and device
CN113379770B (en) Construction method of nasopharyngeal carcinoma MR image segmentation network, image segmentation method and device
CN114612484B (en) Retina OCT image segmentation method based on unsupervised learning
CN114529718B (en) Construction method of extranodal nasal NK/T cell lymphoma diagnosis and prognosis model
CN116439691A (en) Joint activity detection method based on artificial intelligence and related equipment
CN109544369B (en) Tuberculosis authentication method based on data processing and related equipment
CN111445406B (en) Low-dose CT picture quality improvement method, system and equipment
CN110874614B (en) Brain image classification method, computer device, and readable storage medium
CN113052930A (en) Chest DR dual-energy digital subtraction image generation method
CN111080572B (en) White matter high signal positioning method, white matter high signal positioning device, white matter high signal positioning equipment and storage medium
WO2021107098A1 (en) Document creation assistance device, document creation assistance method, and document creation assistance program
CN115148341B (en) AI structure sketching method and system based on body position recognition
CN117078664B (en) Computer-readable storage medium, ultrasonic image quality evaluation device, and electronic apparatus
CN115619641B (en) FFDM-based breast image processing method, FFDM-based breast image processing system, FFDM-based terminal and FFDM-based breast image processing medium
CN115578285B (en) Mammary gland molybdenum target image detail enhancement method and system

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20211022

RJ01 Rejection of invention patent application after publication