CN111709958A - Visualization method of brain nuclear magnetic resonance abnormal image based on 3D CAM - Google Patents
Visualization method of brain nuclear magnetic resonance abnormal image based on 3D CAM Download PDFInfo
- Publication number
- CN111709958A CN111709958A CN202010577553.9A CN202010577553A CN111709958A CN 111709958 A CN111709958 A CN 111709958A CN 202010577553 A CN202010577553 A CN 202010577553A CN 111709958 A CN111709958 A CN 111709958A
- Authority
- CN
- China
- Prior art keywords
- abnormal
- magnetic resonance
- cam
- nuclear magnetic
- brain
- 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
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 122
- 210000004556 brain Anatomy 0.000 title claims abstract description 42
- 238000005481 NMR spectroscopy Methods 0.000 title claims abstract description 24
- 238000007794 visualization technique Methods 0.000 title claims abstract description 18
- 238000012549 training Methods 0.000 claims abstract description 25
- 238000001514 detection method Methods 0.000 claims abstract description 8
- 238000000034 method Methods 0.000 claims description 12
- 238000005259 measurement Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000011160 research Methods 0.000 abstract description 5
- 238000012800 visualization Methods 0.000 abstract description 5
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004445 quantitative analysis Methods 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000002595 magnetic resonance imaging Methods 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention discloses a visualization method of a brain nuclear magnetic resonance abnormal image based on a 3D CAM, which comprises the following steps: s1, acquiring brain nuclear magnetic resonance abnormal image data of a patient and corresponding abnormal recognition result data as a training sample; s2, constructing an abnormal image recognition model by using a training sample, wherein the abnormal image recognition model is trained by adopting an Ssd _ Grad-CAM model; s3, recognizing the magnetic resonance image of the brain based on the abnormal image recognition model, and marking a position frame and an abnormal recognition result in the abnormal area if the abnormal area exists; and S4, realizing three-dimensional reconstruction of the magnetic resonance image based on MATLAB, and marking a corresponding abnormal region and an abnormal recognition result on the obtained three-dimensional model. The invention can realize the automatic identification and detection and three-dimensional visualization of the brain nuclear magnetic resonance abnormal image, and is convenient for assisting medical researchers to quantitatively analyze and research the brain nuclear magnetic resonance abnormal image.
Description
Technical Field
The invention relates to the technical field of image visualization, in particular to a visualization method of a brain nuclear magnetic resonance abnormal image based on a 3D CAM.
Background
The brain is the most complex organ currently known to human beings, and in order to understand the brain well, many efforts are made, and Magnetic Resonance Imaging (MRI) technology is an important breakthrough.
In recent years, with the rise of machine learning, medical data is used in combination with machine learning more and more, and it is a prerequisite to process the medical data well for effective use.
Three-dimensional visualization of magnetic resonance images in the medical field is a hot problem in current research, and has important applications in diagnostic medicine, surgical planning, simulation and the like. Therefore, the method has important academic significance and application value for the research of three-dimensional visualization of the magnetic resonance image. .
Disclosure of Invention
The invention aims to provide a visualization method of a brain nuclear magnetic resonance abnormal image based on a 3D CAM, which can realize automatic identification detection and three-dimensional visualization of the brain nuclear magnetic resonance abnormal image and is convenient for assisting a medical researcher in quantitative analysis and research of the brain nuclear magnetic resonance abnormal image.
In order to achieve the purpose, the invention adopts the technical scheme that:
the visualization method of the brain nuclear magnetic resonance abnormal image based on the 3D CAM is characterized by comprising the following steps: the method comprises the following steps:
s1, acquiring brain nuclear magnetic resonance abnormal image data of a patient and corresponding abnormal recognition result data as a training sample;
s2, constructing an abnormal image recognition model by using a training sample, wherein the abnormal image recognition model is trained by adopting an Ssd _ Grad-CAM model;
s3, recognizing the magnetic resonance image of the brain based on the abnormal image recognition model, and marking a position frame and an abnormal recognition result in the abnormal area if the abnormal area exists;
and S4, realizing three-dimensional reconstruction of the magnetic resonance image based on MATLAB, and marking a corresponding abnormal region and an abnormal recognition result on the obtained three-dimensional model.
Further, the Ssd _ Grad-CAM model adopts an Ssd target detection algorithm, and Grad-CAM is trained by training samples.
Further, the method also comprises the steps of measuring the size of the abnormal area and acquiring size data.
Further, the method also comprises a step of marking the abnormal area size data at the corresponding position of the abnormal area.
Further, the method also comprises the steps of constructing a three-dimensional model coordinate system and acquiring three-dimensional coordinate data of each vertex of the abnormal area.
Further, during size measurement, automatic threshold segmentation of the finished image is performed by adopting an Otsu algorithm, and then three-dimensional size parameters of the abnormal region are measured based on the length-width ratio of a connected component circumscribed rectangle.
The invention has the following beneficial effects:
the abnormal region in the magnetic resonance image of the brain is automatically identified based on the Ssd _ Grad-CAM model, and the efficiency of identifying the abnormal region can be greatly improved while the accuracy of identifying the abnormal region can be improved.
Three-dimensional reconstruction of the magnetic resonance image is realized based on MATLAB, and corresponding abnormal regions, abnormal recognition results, coordinates of vertexes of the abnormal regions and size data of the abnormal regions are marked on the obtained three-dimensional model, so that quantitative analysis and research of brain nuclear magnetic resonance abnormal images by medical researchers are greatly facilitated.
Drawings
Fig. 1 is a flowchart of a visualization method of a brain magnetic resonance abnormal image based on 3D CAM in embodiment 1 of the present invention.
Fig. 2 is a flowchart of a visualization method for a brain magnetic resonance abnormal image based on 3D CAM in embodiment 2 of the present invention.
Fig. 3 is a flowchart of a visualization method of a brain magnetic resonance abnormal image based on 3D CAM according to embodiment 3 of the present invention.
Fig. 4 is a flowchart of a visualization method of a brain magnetic resonance abnormal image based on 3D CAM according to embodiment 4 of the present invention.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that variations and modifications can be made by persons skilled in the art without departing from the spirit of the invention. All falling within the scope of the present invention.
Example 1
As shown in fig. 1, the visualization method of the brain magnetic resonance abnormal image based on the 3D CAM includes the following steps:
s1, acquiring brain nuclear magnetic resonance abnormal image data of a patient and corresponding abnormal recognition result data as a training sample;
s2, constructing an abnormal image recognition model by using a training sample, wherein the abnormal image recognition model is trained by adopting an Ssd _ Grad-CAM model;
s3, recognizing the magnetic resonance image of the brain based on the abnormal image recognition model, and marking a position frame and an abnormal recognition result in the abnormal area if the abnormal area exists;
and S4, realizing three-dimensional reconstruction of the magnetic resonance image based on MATLAB, and marking a corresponding abnormal region and an abnormal recognition result on the obtained three-dimensional model.
In this embodiment, the Ssd _ Grad-CAM model is obtained by training Grad-CAM with a training sample by using an Ssd target detection algorithm.
Example 2
As shown in fig. 2, the method for visualizing the magnetic resonance abnormal brain image based on the 3D CAM includes the following steps:
s1, acquiring brain nuclear magnetic resonance abnormal image data of a patient and corresponding abnormal recognition result data as a training sample;
s2, constructing an abnormal image recognition model by using a training sample, wherein the abnormal image recognition model is trained by adopting an Ssd _ Grad-CAM model;
s3, recognizing the magnetic resonance image of the brain based on the abnormal image recognition model, and marking a position frame and an abnormal recognition result in the abnormal area if the abnormal area exists;
s4, realizing three-dimensional reconstruction of the magnetic resonance image based on MATLAB, and marking a corresponding abnormal region and an abnormal recognition result on the obtained three-dimensional model;
s5, acquiring the abnormal area size data, and marking the abnormal area size data at the corresponding position of the abnormal area; during size measurement, automatic threshold segmentation of a finished image is performed by adopting an Otsu algorithm, and then three-dimensional size parameters of an abnormal region are measured based on the length-width ratio of a connected component circumscribed rectangle.
In this embodiment, the Ssd _ Grad-CAM model is obtained by training Grad-CAM with a training sample by using an Ssd target detection algorithm.
Example 3
As shown in fig. 3, the visualization method of the brain magnetic resonance abnormal image based on the 3D CAM includes the following steps:
s1, acquiring brain nuclear magnetic resonance abnormal image data of a patient and corresponding abnormal recognition result data as a training sample;
s2, constructing an abnormal image recognition model by using a training sample, wherein the abnormal image recognition model is trained by adopting an Ssd _ Grad-CAM model;
s3, recognizing the magnetic resonance image of the brain based on the abnormal image recognition model, and marking a position frame and an abnormal recognition result in the abnormal area if the abnormal area exists;
s4, realizing three-dimensional reconstruction of the magnetic resonance image based on MATLAB, and marking a corresponding abnormal region and an abnormal recognition result on the obtained three-dimensional model;
and S5, building a three-dimensional model coordinate system, and acquiring three-dimensional coordinate data of each vertex of the abnormal area.
In this embodiment, the Ssd _ Grad-CAM model is obtained by training Grad-CAM with a training sample by using an Ssd target detection algorithm.
Example 4
As shown in fig. 4, the method for visualizing the magnetic resonance abnormal brain image based on the 3D CAM includes the following steps:
s1, acquiring brain nuclear magnetic resonance abnormal image data of a patient and corresponding abnormal recognition result data as a training sample;
s2, constructing an abnormal image recognition model by using a training sample, wherein the abnormal image recognition model is trained by adopting an Ssd _ Grad-CAM model;
s3, recognizing the magnetic resonance image of the brain based on the abnormal image recognition model, and marking a position frame and an abnormal recognition result in the abnormal area if the abnormal area exists;
s4, realizing three-dimensional reconstruction of the magnetic resonance image based on MATLAB, and marking a corresponding abnormal region and an abnormal recognition result on the obtained three-dimensional model;
s5, acquiring the abnormal area size data, and marking the abnormal area size data at the corresponding position of the abnormal area; during size measurement, firstly, an Otsu algorithm is adopted to automatically segment a threshold value of a finished image, and then three-dimensional size parameters of an abnormal region are measured based on the length-width ratio of a connected component circumscribed rectangle;
and S6, building a three-dimensional model coordinate system, and acquiring three-dimensional coordinate data of each vertex of the abnormal area.
In this embodiment, the Ssd _ Grad-CAM model is obtained by training Grad-CAM with a training sample by using an Ssd target detection algorithm.
The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.
Claims (6)
1. The visualization method of the brain nuclear magnetic resonance abnormal image based on the 3DCAM is characterized by comprising the following steps: the method comprises the following steps:
s1, acquiring brain nuclear magnetic resonance abnormal image data of a patient and corresponding abnormal recognition result data as a training sample;
s2, constructing an abnormal image recognition model by using a training sample, wherein the abnormal image recognition model is trained by adopting an Ssd _ Grad-CAM model;
s3, recognizing the magnetic resonance image of the brain based on the abnormal image recognition model, and marking a position frame and an abnormal recognition result in the abnormal area if the abnormal area exists;
and S4, realizing three-dimensional reconstruction of the magnetic resonance image based on MATLAB, and marking a corresponding abnormal region and an abnormal recognition result on the obtained three-dimensional model.
2. The visualization method for the brain nuclear magnetic resonance abnormal image based on the 3D CAM as claimed in claim 1, wherein: the Ssd _ Grad-CAM model is obtained by training Grad-CAM with training samples by adopting an Ssd target detection algorithm.
3. The visualization method for the brain nuclear magnetic resonance abnormal image based on the 3D CAM as claimed in claim 1, wherein: the method also comprises the steps of measuring the size of the abnormal area and acquiring size data.
4. The visualization method for the brain nuclear magnetic resonance abnormal image based on the 3D CAM as claimed in claim 3, characterized in that: the method also comprises a step of marking the abnormal area size data at the corresponding position of the abnormal area.
5. The visualization method for the brain nuclear magnetic resonance abnormal image based on the 3D CAM as claimed in claim 1, wherein: the method also comprises the steps of constructing a three-dimensional model coordinate system and acquiring three-dimensional coordinate data of each vertex of the abnormal area.
6. The visualization method for the brain nuclear magnetic resonance abnormal image based on the 3D CAM as claimed in claim 3, characterized in that: during size measurement, automatic threshold segmentation of a finished image is performed by adopting an Otsu algorithm, and then three-dimensional size parameters of an abnormal region are calculated based on the length-width ratio of a connected component circumscribed rectangle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010577553.9A CN111709958A (en) | 2020-06-22 | 2020-06-22 | Visualization method of brain nuclear magnetic resonance abnormal image based on 3D CAM |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010577553.9A CN111709958A (en) | 2020-06-22 | 2020-06-22 | Visualization method of brain nuclear magnetic resonance abnormal image based on 3D CAM |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111709958A true CN111709958A (en) | 2020-09-25 |
Family
ID=72541555
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010577553.9A Pending CN111709958A (en) | 2020-06-22 | 2020-06-22 | Visualization method of brain nuclear magnetic resonance abnormal image based on 3D CAM |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111709958A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103300862A (en) * | 2013-05-24 | 2013-09-18 | 浙江大学宁波理工学院 | Method for measuring depth and three-dimensional size of nidus tissue by capsule endoscope |
| CN103886603A (en) * | 2014-03-31 | 2014-06-25 | 西北工业大学 | Left ventricle nuclear magnetic resonance image segmentation and three-dimensional reconstruction method |
| US20160065855A1 (en) * | 2013-05-07 | 2016-03-03 | Commissariat A L'energie Atomique Et Aux Energies Alternatives | Method for controlling a graphical interface for displaying images of a three-dimensional object |
| CN108986088A (en) * | 2018-07-10 | 2018-12-11 | 安徽师范大学 | A kind of image based on MATLAB automatically extracts optimization method and equipment |
| CN110148452A (en) * | 2019-05-07 | 2019-08-20 | 东软医疗系统股份有限公司 | A kind of image rendering method and device |
| CN110176001A (en) * | 2019-06-03 | 2019-08-27 | 浙江大学 | A kind of high iron catenary insulator breakage accurate positioning method based on Grad-CAM algorithm |
| CN110910366A (en) * | 2019-11-18 | 2020-03-24 | 湖北工业大学 | Visualization method of brain nuclear magnetic resonance abnormal image based on 3D CAM |
| CN111243042A (en) * | 2020-02-28 | 2020-06-05 | 浙江德尚韵兴医疗科技有限公司 | Ultrasonic thyroid nodule benign and malignant characteristic visualization method based on deep learning |
-
2020
- 2020-06-22 CN CN202010577553.9A patent/CN111709958A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160065855A1 (en) * | 2013-05-07 | 2016-03-03 | Commissariat A L'energie Atomique Et Aux Energies Alternatives | Method for controlling a graphical interface for displaying images of a three-dimensional object |
| CN103300862A (en) * | 2013-05-24 | 2013-09-18 | 浙江大学宁波理工学院 | Method for measuring depth and three-dimensional size of nidus tissue by capsule endoscope |
| CN103886603A (en) * | 2014-03-31 | 2014-06-25 | 西北工业大学 | Left ventricle nuclear magnetic resonance image segmentation and three-dimensional reconstruction method |
| CN108986088A (en) * | 2018-07-10 | 2018-12-11 | 安徽师范大学 | A kind of image based on MATLAB automatically extracts optimization method and equipment |
| CN110148452A (en) * | 2019-05-07 | 2019-08-20 | 东软医疗系统股份有限公司 | A kind of image rendering method and device |
| CN110176001A (en) * | 2019-06-03 | 2019-08-27 | 浙江大学 | A kind of high iron catenary insulator breakage accurate positioning method based on Grad-CAM algorithm |
| CN110910366A (en) * | 2019-11-18 | 2020-03-24 | 湖北工业大学 | Visualization method of brain nuclear magnetic resonance abnormal image based on 3D CAM |
| CN111243042A (en) * | 2020-02-28 | 2020-06-05 | 浙江德尚韵兴医疗科技有限公司 | Ultrasonic thyroid nodule benign and malignant characteristic visualization method based on deep learning |
Non-Patent Citations (4)
| Title |
|---|
| E. ENNADIFI, S. LARABA, D. VINCKE, B. MERCATORIS AND B. GOSSELIN: "Wheat Diseases Classification and Localization Using Convolutional Neural Networks and GradCAM Visualization" * |
| SELVARAJU R R, COGSWELL M, DAS A, ET AL: "Grad-cam: Visual explanations from deep networks via gradient-based localization" * |
| ZHOU B, KHOSLA A, LAPEDRIZA A, ET AL: "Learning deep features for discriminative localization" * |
| 黄开云, 郑政, 杨柳: "基于MATLAB的三维超声成像及图像处理" * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shi et al. | Review of artificial intelligence techniques in imaging data acquisition, segmentation, and diagnosis for COVID-19 | |
| WO2019052062A1 (en) | Medical image report generation system and method based on artificial intelligence | |
| CN110490851B (en) | Mammary gland image segmentation method, device and system based on artificial intelligence | |
| CN111539956B (en) | Cerebral hemorrhage automatic detection method based on brain auxiliary image and electronic medium | |
| El-Baz et al. | Lung imaging data analysis | |
| CN110717905B (en) | Brain image detection method, computer device, and storage medium | |
| Jannin et al. | Validation in medical image processing. | |
| CN111311703A (en) | Electrical impedance tomography image reconstruction method based on deep learning | |
| Elnakib et al. | Dyslexia diagnostics by 3-D shape analysis of the corpus callosum | |
| CN112365980A (en) | Brain tumor multi-target point auxiliary diagnosis and prospective treatment evolution visualization method and system | |
| CN110298820A (en) | Image analysis methods, computer equipment and storage medium | |
| US20240127447A1 (en) | Quantitative framework for the diagnostic, prognostic, and therapeutic evaluation of spinal cord diseases | |
| Li et al. | BrainK for structural image processing: creating electrical models of the human head | |
| Alam et al. | Evaluation of medical image registration techniques based on nature and domain of the transformation | |
| Xie et al. | Interpretable diagnosis for whole-slide melanoma histology images using convolutional neural network | |
| JP2022546344A (en) | Image processing for stroke feature acquisition | |
| CN114065825B (en) | Brain magnetic MEG source positioning method based on structural similarity | |
| Zeng et al. | TUSPM-NET: A multi-task model for thyroid ultrasound standard plane recognition and detection of key anatomical structures of the thyroid | |
| CN106510708A (en) | Framework for Abnormality Detection in Multi-Contrast Brain Magnetic Resonance Data | |
| CN111709958A (en) | Visualization method of brain nuclear magnetic resonance abnormal image based on 3D CAM | |
| CN110570425A (en) | Lung nodule analysis method and device based on deep reinforcement learning algorithm | |
| CN115035086A (en) | Intelligent tuberculosis skin test screening and analyzing method and device based on deep learning | |
| CN113469942B (en) | CT image lesion detection method | |
| JP2023551131A (en) | Guided acquisition of 3D representations of anatomical structures | |
| CN113469987A (en) | Dental X-ray image lesion area positioning system based on deep learning |
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: 20200925 |
|
| RJ01 | Rejection of invention patent application after publication |