CN107330267A - Utilize the white matter fiber brain map construction method of diffusion tensor medical image - Google Patents
Utilize the white matter fiber brain map construction method of diffusion tensor medical image Download PDFInfo
- Publication number
- CN107330267A CN107330267A CN201710509098.7A CN201710509098A CN107330267A CN 107330267 A CN107330267 A CN 107330267A CN 201710509098 A CN201710509098 A CN 201710509098A CN 107330267 A CN107330267 A CN 107330267A
- Authority
- CN
- China
- Prior art keywords
- image
- carried out
- tensor
- feature
- registration
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/32—Determination of transform parameters for the alignment of images, i.e. image registration using correlation-based methods
-
- 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]
- G06T2207/10092—Diffusion tensor magnetic resonance imaging [DTI]
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
Abstract
Using the white matter fiber brain map construction method of diffusion tensor, including：Input original diffusion tensor image (101)；To dispersion tensor image carry out image preprocessing (106), including Data Format Transform (102), be converted to 4D images (103), calculate anisotropic value (104), operation (105) of boning；Using registering (107) technology, using reflection transformation (108) and the differomorphism mapping algorithm (109) of large deformation, same standardised space is arrived into the dispersion tensor image registration (107) of different subjects；Diffusion-weighted imaging analysis (113) is carried out, (110) is analyzed Deformation Field using tensor model, carries out diffusion-weighted imaging redirection (111) and dispersion tensor estimation (112)；Two white matter of brain collection of illustrative plates (115) are constructed for patient and normal person；Machine learning (114) is carried out, including：Feature extraction (116) and feature selecting (117) are carried out to subject picture, data set is divided into training set and test set, control collection of illustrative plates carries out machine learning (118), training generation grader (121).
Description
Background technology
The white matter fiber brain map construction method of diffusion tensor medical image is the present invention relates to the use of, for brain disease
Sick clinical medicine auxiliary diagnosis.
Background technology
Diffusion tensor technology is widely used in probing into live body brain white as a kind of special MR imaging method
The design feature of matter fiber.The imaging technique describes cerebral white matter indirectly by detecting the diffusion property of brain hydrone
The geometric shape of fibre structure and space trend, this research developed to cerebral white matter and the research of some neurogenic diseases rise
Very important effect is arrived.
For public brain map template conventional at present, the structural information on the white matter of brain is more lacked.This
It is because in traditional magnetic resonance imaging image, white matter information is more single.Traditional magnetic resonance imaging template institute body
Existing anatomic information is single gray value information, and it is extremely difficult to describe the white matter region of characteristic for identification.Disperse
Tensor imaging is used as a kind of new resonance configurations so that we can carry out simulation to fibre bundle in white matter and rebuild and visual
Change.This image comprising directional information has started a kind of new mode to set up the coordinate system of tensor space to study pathology
Or white matter Structure and Function.Compared to other invasive methods such as PET images, diffusion tensor is to human body without any pair
Effect, with unrivaled superiority, therefore carries out the foundation of the brain map based on diffusion tensor technology, for brain
The clinical assistant diagnosis of disease has larger meaning, and result of study also there is larger Clinical practice to be worth and be widely applied
Prospect.
The content of the invention
Digitlization statistics collection of illustrative plates is to calculate a highly important branch in Nervous System Anatomy research framework.By to big rule
Mould normal population medical image sample data set carries out the study and modeling in statistical significance, and normal population dissection knot is contained in formation
Structure common feature, and the statistics collection of illustrative plates of the digitlization with certain generalization ability, are detection unknown individual and normal population structure shape
The basis of state otherness.Research, which creates three-dimensional diffusion tensor digitlization statistics collection of illustrative plates, has highly important meaning
The present invention is by setting up a kind of method that particularity to dispersion tensor image is analyzed, to original dispersion tensor
Imaging data carries out data conversion, pretreatment and medical figure registration, it is proposed that build the digitlization based on diffusion tensor
A solution of collection of illustrative plates is counted, and using machine learning method to learning for the identification and diagnosis that are tested classification,
So as to reach the purpose for realizing auxiliary diagnosis.
Utilize the white matter fiber brain map construction method of diffusion tensor, it is characterised in that including：
Input original diffusion tensor image；
Carry out image preprocessing to dispersion tensor image, including Data Format Transform, be converted to four-dimensional image, calculate every
Different in nature value, operation of boning；
Using registration technique by the image registration of different subjects to same standardised space, the wherein registration is used instead
Penetrate the differomorphism mapping algorithm of conversion and large deformation；
Diffusion-weighted imaging analysis is carried out, including：
Deformation Field is analyzed using tensor model,
Diffusion tensor is carried out to redirect and dispersion tensor estimation,
Two white matter of brain collection of illustrative plates are constructed for patient and normal person.
Brief description of the drawings：
Fig. 1 is the stream of the auxiliary diagnosis construction method of utilization white matter fiber brain map according to an embodiment of the invention
Cheng Tu.
Fig. 2 is the flow chart of one embodiment of the data prediction in Fig. 1 embodiment.
Fig. 3 is the flow chart of one embodiment of the image registration in Fig. 1 embodiment.
Fig. 4 is the flow chart of one embodiment of the machine-learning process in Fig. 1 embodiment.
Embodiment
Fig. 1 is the stream of the auxiliary diagnosis construction method of utilization white matter fiber brain map according to an embodiment of the invention
Cheng Tu.
As shown in figure 1, the auxiliary diagnosis construction method bag of the utilization white matter fiber brain map of one embodiment of the present of invention
Include：
Input original diffusion tensor image (101)；
Image preprocessing (106) is carried out to dispersion tensor image, including：
Data Format Transform (102),
4D images (103) are converted to,
Anisotropic value (104) is calculated,
Bone (105)；
Same standardised space is arrived into the DTI image registrations (107) of different subjects using registering (107) technology, wherein originally
The registration (107) of embodiment uses the differomorphism mapping algorithm (109) of affine transformation (108) and large deformation；
Diffusion-weighted imaging analysis (113) is carried out, (110), progress are analyzed Deformation Field using tensor model
Diffusion-weighted imaging redirects (111) and dispersion tensor estimation (112)；
Two white matter of brain collection of illustrative plates (115) are constructed for patient and normal person；
Machine learning (114) is carried out with machine Learning Theory and algorithm, including feature extraction is carried out to subject picture
(116) data set and feature selecting (117), is divided into training set and test set, and compares collection of illustrative plates and carries out machine learning (118),
Training generation grader (121), for auxiliary diagnosis decision-making.
In auxiliary diagnosis, for a new tested individual sample (119), identical is carried out to its dispersion tensor image
Pretreatment and registration, make it be alignd with brain map templatespace (120), and are classified using grader, export auxiliary diagnosis
As a result (122), judge whether this subject sample is ill, that is, realizes auxiliary diagnosis (123).
Fig. 2 is the flow chart of one embodiment of the data prediction in Fig. 1 embodiment.
As shown in Fig. 2 the pretreatment (106) of the embodiment according to the figure includes：
Image format conversion (201), by original DICOM data conversions into the NIfTI forms available for analysis；
Three-dimensional dispersion tensor image is converted into four-dimensional image (202), in order to subsequently use and analyze；
To being tested into the dynamic correction (203) of wardrobe, with the head existed to subject sample when dispersion tensor image is scanned
It is dynamic to compensate；
Vortex correction (204) is carried out to image, with time deviation present in compensated scanning process；
Anisotropy value (FA) (205) is calculated, and with FA>0.2 makes brain area template (206)；
Operation (207) of boning is carried out to subject image.
Fig. 3 is the flow chart of one embodiment of the image registration in Fig. 1 embodiment.
As shown in figure 3, the image registration operation of the embodiment according to the figure includes：
Select reference picture (301) and image subject to registration (302)；Image preprocessing is carried out to image subject to registration；
Feature extraction and matching (304) is carried out, i.e., mutually relevant and useful feature information is extracted from two images,
These information are all the essential characteristics of image, including gray scale, characteristic point, edge contour or texture etc.；
Similarity measurement (305) is carried out, for judging whether two images are perfectly aligned, to weigh after spatial alternation
It is consistent that image subject to registration and reference picture realize space in which kind of degree, and image registration problem is thus converted a space
The Solve problems of transformation parameter；
Repeatedly optimize search (307), spatial alternation (308) and image interpolation (306) so that be defined on these changes
The similarity measurement changed in parameter gradually tends to optimal value, even if obtaining spatial transformation parameter constantly approaches optimal registration position；
When iteration tends to be optimal, registration result (309) is exported.
Fig. 4 is the flow chart of one embodiment of the machine-learning process in Fig. 1 embodiment.
As shown in figure 4, the machine-learning process of the embodiment according to the figure includes：
For brain diseases patient and the class sample (401) of normal person two, first to input sample (dispersion tensor image) simultaneously
Carry out initial characteristicses distribution situation analysis (402)；
By the two class diffusion tensor white matter of brain collection of illustrative plates of the normal person generated before and patient, as standard sample database,
And using white matter of brain collection of illustrative plates carry out feature extraction (403), including by spatial relation characteristics (415), edge feature (416),
Shape facility (417), textural characteristics (418) and gray feature (419) carry out feature extraction, form the characteristic set of collection of illustrative plates
(420)；
Feature extraction to single-subject dispersion tensor image (414) is also carried out in this way, forms single-subject special
(421) are closed in collection；
To the dispersion tensor image of single-subject also from spatial relation characteristics (415), edge feature (416), shape facility
(417), textural characteristics (418) and gray feature (419) carry out feature extraction, so as to obtain the characteristic set of single-subject；
(404) are normalized, brain diseases diagnosis problem is converted into a classification problem of machine learning two；
Characteristic validity analysis (405) is carried out, to verify the validity of feature；
Carry out feature selecting (406)；
Carry out combinations of features and conversion (407)；
For the feature, the common classification method of machine learning, such as SVMs, decision tree, deep learning are used
Convolutional neural networks, classified, and by building grader, training grader, checking grader, obtain one automatic point
Class model (408)；
Tuning processing is carried out, including is iteratively repeated characteristic validity analysis (409), hyper parameter adjustment (410), whole process
Signature analysis, sample analysis (411), with lift scheme accuracy, and improve generalization ability；
Export final mask (412).
For new subject sample, above-mentioned series of preprocessing is repeated, feature is extracted, substitutes into model, it is possible to quickly sentence
Whether breaking, it is ill this subject, and provides fiducial probability.
Claims (6)
1. utilize the white matter fiber brain map construction method of diffusion tensor, it is characterised in that including：
Input original diffusion tensor image (101)；
Image preprocessing (106) is carried out to dispersion tensor image, including Data Format Transform (102), is converted to four-dimensional image
(103) anisotropic value (104), operation (105) of boning, are calculated；
Using registering (107) technology by the image registration of different subjects to same standardised space, the wherein registration is used
Reflection transformation (108) and the differomorphism mapping algorithm (109) of large deformation；
Diffusion-weighted imaging analysis (113) is carried out, including：
(110) are analyzed Deformation Field using tensor model,
Diffusion tensor redirection (111) and dispersion tensor estimation (112) are carried out,
Two white matter of brain collection of illustrative plates (115) are constructed for patient and normal person.
2. according to the method described in claim 1, it is characterised in that further comprise：
Carry out machine learning (114) with machine Learning Theory and algorithm, including subject picture is carried out feature extraction (116) and
Feature selecting (117),
Data set is divided into training set and test set, and compares collection of illustrative plates and carries out machine learning (118), training generation grader
(121)。
3. according to the method described in claim 1, it is characterised in that image preprocessing (106) includes：
Image format conversion (201), the NIfTI forms available for analysis are converted into by original DICOM；
Three-dimensional dispersion tensor image is converted into four-dimensional image, in order to subsequently use and analyze；
To being tested into the dynamic correction of wardrobe, mended so that the head existed to subject sample when diffusion tensor is scanned is dynamic
Repay；
Vortex correction is carried out to image, with time deviation present in compensated scanning process；
Anisotropy value FA is calculated, and with FA>0.2 makes brain area template；
Operation of boning is carried out to subject image.
4. according to the method described in claim 1, it is characterised in that diffusion tensor registration includes：
Select reference picture and image subject to registration；
Image preprocessing is carried out to image subject to registration；
Feature extraction and matching is carried out, including mutually relevant and useful feature information is extracted from two images, these are believed
Breath is all the essential characteristic of image, including gray scale, characteristic point, edge contour or texture；
Similarity measurement is carried out, to judge whether two images are perfectly aligned, to weigh the image subject to registration after spatial alternation
Realize that space is consistent in which kind of degree with reference picture, image registration problem is thus converted spatial transformation parameter
Solve problems；
Repeatedly optimize search, spatial alternation and image interpolation so that be defined on the similarity measurements in these transformation parameters
Amount gradually tends to optimal value, even if obtaining spatial transformation parameter constantly approaches optimal registration position；
When iteration tends to be optimal, registration result is exported.
5. method according to claim 2, it is characterised in that machine learning includes：
For brain diseases patient and the class sample (401) of normal person two, the dispersion tensor image to input sample and carry out first
Initial characteristicses distribution situation analyzes (402)；
Using the two class diffusion tensor white matter of brain collection of illustrative plates of the normal person generated before and patient as standard sample database, and utilize
White matter of brain collection of illustrative plates carries out feature extraction (403), including by special to spatial relation characteristics (415), edge feature (416), shape
Levy (417), textural characteristics (418) and gray feature (419) and carry out feature extraction, form the characteristic set (420) of collection of illustrative plates；
To the dispersion tensor image of single-subject, also from spatial relation characteristics (415), edge feature (416), shape facility
(417), textural characteristics (418) and gray feature (419) carry out feature extraction, so as to obtain the characteristic set of single-subject；
(404) are normalized, brain diseases diagnosis problem is converted into a classification problem of machine learning two；
Characteristic validity analysis (405) is carried out, to verify the validity of feature；
Carry out feature selecting (406)；
Carry out combinations of features and conversion (407)；
For the feature, classified using the common classification method of machine learning, and classified by building grader, training
Device, checking grader, obtain an automatic disaggregated model (408)；
Tuning processing is carried out, including is iteratively repeated characteristic validity analysis (409), hyper parameter adjustment (410), whole process feature
Analysis, sample analysis (411), with lift scheme accuracy, and improve generalization ability,
Export final mask (412).
6. method according to claim 5, it is characterised in that the common classification method includes SVMs, decision-making
Tree, the convolutional neural networks of deep learning.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710509098.7A CN107330267A (en) | 2017-06-28 | 2017-06-28 | Utilize the white matter fiber brain map construction method of diffusion tensor medical image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710509098.7A CN107330267A (en) | 2017-06-28 | 2017-06-28 | Utilize the white matter fiber brain map construction method of diffusion tensor medical image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107330267A true CN107330267A (en) | 2017-11-07 |
Family
ID=60198261
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710509098.7A Pending CN107330267A (en) | 2017-06-28 | 2017-06-28 | Utilize the white matter fiber brain map construction method of diffusion tensor medical image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107330267A (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108133509A (en) * | 2017-12-12 | 2018-06-08 | 重庆花椒科技有限公司 | Method and apparatus based on four-dimensional color ultrasound figure structure 3D models |
| CN108652630A (en) * | 2018-04-13 | 2018-10-16 | 廉江市人民医院 | A kind of neurosurgery cerebral operations auxiliary machine |
| CN109035265A (en) * | 2018-09-26 | 2018-12-18 | 重庆邮电大学 | A kind of novel structure brain connection map construction method |
| CN109389587A (en) * | 2018-09-26 | 2019-02-26 | 上海联影智能医疗科技有限公司 | A kind of medical image analysis system, device and storage medium |
| CN109799472A (en) * | 2019-03-08 | 2019-05-24 | 沈阳工业大学 | A kind of magnetic resonance eddy current compensation method based on deep learning |
| CN110288593A (en) * | 2019-07-01 | 2019-09-27 | 山东省肿瘤防治研究院（山东省肿瘤医院） | The post-processing approach and system of diffusion weighted images |
| CN110415228A (en) * | 2019-07-24 | 2019-11-05 | 上海联影医疗科技有限公司 | Nerve fibre method for tracing, magnetic resonance system and storage medium |
| CN110458869A (en) * | 2019-05-10 | 2019-11-15 | 珠海慧脑云计算有限公司 | A kind of method for registering and its system of newborn's magnetic resonance configurations brain image |
| CN110555835A (en) * | 2019-09-04 | 2019-12-10 | 郑州大学 | brain slice image region division method and device |
| CN110833414A (en) * | 2019-11-28 | 2020-02-25 | 广州中医药大学第一附属医院 | Multi-modal molecular imaging strategy of radioactive brain injury biomarker after nasopharyngeal carcinoma radiotherapy |
| CN110993094A (en) * | 2019-11-19 | 2020-04-10 | 中国科学院深圳先进技术研究院 | Intelligent auxiliary diagnosis method and terminal based on medical images |
| CN111354001A (en) * | 2018-12-20 | 2020-06-30 | 西门子医疗系统有限公司 | Brain tumor image segmentation method, device and storage medium |
| CN111407277A (en) * | 2020-03-06 | 2020-07-14 | 中国科学院武汉物理与数学研究所 | Magnetic resonance perfusion-diffusion image registration method for acute ischemic stroke |
| US20200275857A1 (en) * | 2019-03-01 | 2020-09-03 | Siemens Healthcare Gmbh | Tumor Tissue Characterization using Multi-Parametric Magnetic Resonance Imaging |
| CN112614126A (en) * | 2020-12-31 | 2021-04-06 | 中国科学院自动化研究所 | Magnetic resonance image brain region dividing method, system and device based on machine learning |
| CN113222001A (en) * | 2021-05-07 | 2021-08-06 | 天津医科大学 | Construction method and application of morphological fusion classification index of neural image marker |
| CN113221952A (en) * | 2021-04-13 | 2021-08-06 | 山东师范大学 | Multi-center brain diffusion tensor imaging graph classification method and system |
| CN115082586A (en) * | 2022-07-12 | 2022-09-20 | 中国科学院自动化研究所 | Group prior guided thalamus individualized atlas drawing method based on deep learning |
| CN116523857A (en) * | 2023-04-21 | 2023-08-01 | 首都医科大学附属北京友谊医院 | Hearing state prediction device and method based on diffusion tensor image |
| US11969239B2 (en) * | 2020-01-15 | 2024-04-30 | Siemens Healthineers Ag | Tumor tissue characterization using multi-parametric magnetic resonance imaging |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7355403B2 (en) * | 2005-01-27 | 2008-04-08 | Siemens Medical Solutions Usa, Inc. | Noise reduction in diffusion tensor imaging data using bayesian methods |
| CN104361318A (en) * | 2014-11-10 | 2015-02-18 | 中国科学院深圳先进技术研究院 | Disease diagnosis auxiliary system and disease diagnosis auxiliary method both based on diffusion tensor imaging technology |
-
2017
- 2017-06-28 CN CN201710509098.7A patent/CN107330267A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7355403B2 (en) * | 2005-01-27 | 2008-04-08 | Siemens Medical Solutions Usa, Inc. | Noise reduction in diffusion tensor imaging data using bayesian methods |
| CN104361318A (en) * | 2014-11-10 | 2015-02-18 | 中国科学院深圳先进技术研究院 | Disease diagnosis auxiliary system and disease diagnosis auxiliary method both based on diffusion tensor imaging technology |
Non-Patent Citations (2)
| Title |
|---|
| 刘伟: "人脑三维弥散张量影像数字化统计图谱研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
| 李慧卓: "基于集成学习的多模态AD辅助诊断模型研究", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 * |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108133509A (en) * | 2017-12-12 | 2018-06-08 | 重庆花椒科技有限公司 | Method and apparatus based on four-dimensional color ultrasound figure structure 3D models |
| CN108652630A (en) * | 2018-04-13 | 2018-10-16 | 廉江市人民医院 | A kind of neurosurgery cerebral operations auxiliary machine |
| CN109035265A (en) * | 2018-09-26 | 2018-12-18 | 重庆邮电大学 | A kind of novel structure brain connection map construction method |
| CN109389587A (en) * | 2018-09-26 | 2019-02-26 | 上海联影智能医疗科技有限公司 | A kind of medical image analysis system, device and storage medium |
| CN109035265B (en) * | 2018-09-26 | 2020-10-27 | 重庆邮电大学 | Structural brain connection map construction method |
| CN111354001B (en) * | 2018-12-20 | 2024-02-02 | 西门子医疗系统有限公司 | Brain tumor image segmentation method, device and storage medium |
| CN111354001A (en) * | 2018-12-20 | 2020-06-30 | 西门子医疗系统有限公司 | Brain tumor image segmentation method, device and storage medium |
| US20200275857A1 (en) * | 2019-03-01 | 2020-09-03 | Siemens Healthcare Gmbh | Tumor Tissue Characterization using Multi-Parametric Magnetic Resonance Imaging |
| CN111640118B (en) * | 2019-03-01 | 2024-03-01 | 西门子医疗有限公司 | Tumor tissue characterization using multiparameter magnetic resonance imaging |
| CN111640118A (en) * | 2019-03-01 | 2020-09-08 | 西门子医疗有限公司 | Tumor tissue characterization using multi-parameter magnetic resonance imaging |
| CN109799472A (en) * | 2019-03-08 | 2019-05-24 | 沈阳工业大学 | A kind of magnetic resonance eddy current compensation method based on deep learning |
| CN109799472B (en) * | 2019-03-08 | 2022-06-28 | 沈阳工业大学 | Magnetic resonance eddy current compensation method based on deep learning |
| CN110458869A (en) * | 2019-05-10 | 2019-11-15 | 珠海慧脑云计算有限公司 | A kind of method for registering and its system of newborn's magnetic resonance configurations brain image |
| CN110458869B (en) * | 2019-05-10 | 2021-10-19 | 珠海慧脑云计算有限公司 | Registration method and system for brain image of magnetic resonance structure of newborn |
| CN110288593A (en) * | 2019-07-01 | 2019-09-27 | 山东省肿瘤防治研究院（山东省肿瘤医院） | The post-processing approach and system of diffusion weighted images |
| CN110415228A (en) * | 2019-07-24 | 2019-11-05 | 上海联影医疗科技有限公司 | Nerve fibre method for tracing, magnetic resonance system and storage medium |
| CN110555835A (en) * | 2019-09-04 | 2019-12-10 | 郑州大学 | brain slice image region division method and device |
| CN110555835B (en) * | 2019-09-04 | 2022-12-02 | 郑州大学 | Brain slice image region division method and device |
| CN110993094A (en) * | 2019-11-19 | 2020-04-10 | 中国科学院深圳先进技术研究院 | Intelligent auxiliary diagnosis method and terminal based on medical images |
| CN110833414B (en) * | 2019-11-28 | 2021-11-02 | 广州中医药大学第一附属医院 | Multi-modal molecular imaging method of radioactive brain injury biomarker |
| CN110833414A (en) * | 2019-11-28 | 2020-02-25 | 广州中医药大学第一附属医院 | Multi-modal molecular imaging strategy of radioactive brain injury biomarker after nasopharyngeal carcinoma radiotherapy |
| US11969239B2 (en) * | 2020-01-15 | 2024-04-30 | Siemens Healthineers Ag | Tumor tissue characterization using multi-parametric magnetic resonance imaging |
| CN111407277A (en) * | 2020-03-06 | 2020-07-14 | 中国科学院武汉物理与数学研究所 | Magnetic resonance perfusion-diffusion image registration method for acute ischemic stroke |
| CN111407277B (en) * | 2020-03-06 | 2022-04-26 | 中国科学院武汉物理与数学研究所 | Magnetic resonance perfusion-diffusion image registration method for acute ischemic stroke |
| CN112614126A (en) * | 2020-12-31 | 2021-04-06 | 中国科学院自动化研究所 | Magnetic resonance image brain region dividing method, system and device based on machine learning |
| CN113221952A (en) * | 2021-04-13 | 2021-08-06 | 山东师范大学 | Multi-center brain diffusion tensor imaging graph classification method and system |
| CN113221952B (en) * | 2021-04-13 | 2023-09-15 | 山东师范大学 | Multi-center brain diffusion tensor imaging image classification method and system |
| CN113222001A (en) * | 2021-05-07 | 2021-08-06 | 天津医科大学 | Construction method and application of morphological fusion classification index of neural image marker |
| CN115082586A (en) * | 2022-07-12 | 2022-09-20 | 中国科学院自动化研究所 | Group prior guided thalamus individualized atlas drawing method based on deep learning |
| CN116523857B (en) * | 2023-04-21 | 2023-12-26 | 首都医科大学附属北京友谊医院 | Hearing state prediction device and method based on diffusion tensor image |
| CN116523857A (en) * | 2023-04-21 | 2023-08-01 | 首都医科大学附属北京友谊医院 | Hearing state prediction device and method based on diffusion tensor image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107330267A (en) | Utilize the white matter fiber brain map construction method of diffusion tensor medical image | |
| CN107016395A (en) | Primary brain lymthoma and the discrimination method of glioblastoma based on rarefaction representation system | |
| Wu et al. | Registration of longitudinal brain image sequences with implicit template and spatial–temporal heuristics | |
| Liu et al. | Patch-based augmentation of Expectation–Maximization for brain MRI tissue segmentation at arbitrary age after premature birth | |
| Zhu et al. | Discovering dense and consistent landmarks in the brain | |
| US20240127447A1 (en) | Quantitative framework for the diagnostic, prognostic, and therapeutic evaluation of spinal cord diseases | |
| Liu et al. | An enhanced multi-modal brain graph network for classifying neuropsychiatric disorders | |
| CN111932522B (en) | Alzheimer's disease classifier based on brain imaging big data deep learning | |
| CN111488934A (en) | Brain image data processing method, storage medium, computer device and apparatus | |
| Yerukalareddy et al. | Brain tumor classification based on mr images using GAN as a pre-trained model | |
| Mostapha et al. | Towards non-invasive image-based early diagnosis of autism | |
| Ramana | Alzheimer disease detection and classification on magnetic resonance imaging (MRI) brain images using improved expectation maximization (IEM) and convolutional neural network (CNN) | |
| Yang et al. | Diagnosis of Parkinson’s disease based on 3D ResNet: The frontal lobe is crucial | |
| CN107590806A (en) | A kind of detection method and system based on brain medical imaging | |
| Noor et al. | Automated segmentation and classification technique for brain stroke | |
| Qi et al. | Age estimation from MR images via 3D convolutional neural network and densely connect | |
| Schilling et al. | AI in MRI: A case for grassroots deep learning | |
| CN115937129A (en) | Method and device for processing left-right half-brain relation based on multi-modal magnetic resonance image | |
| Avila et al. | Inference of an extended short fiber bundle atlas using sulcus-based constraints for a diffeomorphic inter-subject alignment | |
| TWI477798B (en) | Method of automatically analyzing brain fiber tracts information | |
| Cheng et al. | Low rank self-calibrated brain network estimation and autoweighted centralized multi-task learning for early mild cognitive impairment diagnosis | |
| Commowick et al. | Non-local robust detection of DTI white matter differences with small databases | |
| Orasanu et al. | Tensor spectral matching of diffusion weighted images | |
| Kawamoto et al. | Simultaneous Learning of Erector Spinae Muscles for Automatic Segmentation of Site-Specific Skeletal Muscles in Body CT Images | |
| Zhao et al. | HS–GS: a method for multicenter MR image standardization |
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 |