CN111341443A - Ultrasonic thyroid nodule intelligent evaluation method based on deep learning - Google Patents
Ultrasonic thyroid nodule intelligent evaluation method based on deep learning Download PDFInfo
- Publication number
- CN111341443A CN111341443A CN202010155463.0A CN202010155463A CN111341443A CN 111341443 A CN111341443 A CN 111341443A CN 202010155463 A CN202010155463 A CN 202010155463A CN 111341443 A CN111341443 A CN 111341443A
- Authority
- CN
- China
- Prior art keywords
- deep learning
- thyroid
- target
- nodule
- data
- 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
- 208000009453 Thyroid Nodule Diseases 0.000 title claims abstract description 20
- 208000024770 Thyroid neoplasm Diseases 0.000 title claims abstract description 18
- 238000013135 deep learning Methods 0.000 title claims abstract description 15
- 238000011156 evaluation Methods 0.000 title claims abstract description 7
- 210000001685 thyroid gland Anatomy 0.000 claims abstract description 15
- 238000000034 method Methods 0.000 claims abstract description 13
- 238000002592 echocardiography Methods 0.000 claims abstract description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 9
- 238000003745 diagnosis Methods 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 9
- 238000002604 ultrasonography Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 3
- 238000011176 pooling Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 238000004880 explosion Methods 0.000 description 2
- 206010061218 Inflammation Diseases 0.000 description 1
- 206010028980 Neoplasm Diseases 0.000 description 1
- 208000024799 Thyroid disease Diseases 0.000 description 1
- 230000005784 autoimmunity Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007850 degeneration Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004054 inflammatory process Effects 0.000 description 1
- 230000000152 swallowing effect Effects 0.000 description 1
- 201000002510 thyroid cancer Diseases 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- 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
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
-
- 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
- G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
-
- 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
- G06T2207/10132—Ultrasound image
-
- 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
- G06T2207/30096—Tumor; Lesion
Abstract
The invention relates to the technical field of medical AI. The invention discloses an ultrasonic thyroid nodule intelligent evaluation method based on deep learning, which adopts an automatic end-to-end deep learning method to identify a thyroid ultrasonic image, obtains all information of a thyroid nodule, obtains the most important characteristic indexes of the nodule except the position of the nodule, and comprises the information of components, echoes, shapes, boundaries and punctiform strong echoes, and visually gives the final judgment and suggestion.
Description
Technical Field
The invention belongs to the technical field of medical AI (artificial intelligence), and particularly relates to an ultrasonic thyroid nodule intelligent evaluation method based on deep learning.
Background
Thyroid nodule is a lump in the thyroid gland, can move up and down with the thyroid gland along with swallowing action, is a common clinical disease and can be caused by various causes. Clinically, various thyroid diseases such as thyroid degeneration, inflammation, autoimmunity, neoplasms and the like can be expressed as nodules. Thyroid nodules can be single-shot or multiple-shot, and multiple nodules have higher morbidity than single nodules, but the incidence rate of thyroid cancer of single nodules is higher.
There are many ways to examine thyroid nodules, and among them, ultrasonic diagnosis is a relatively common examination method, and is significant in distinguishing the size of the nodule, identifying the nodule position, and guiding positioning puncture. The existing ultrasonic diagnosis modes are that ultrasonic scanning is carried out on the thyroid of a patient by adopting ultrasonic waves to form a thyroid ultrasonic picture, then a doctor carries out manual identification and judgment on the thyroid ultrasonic picture, and the existing ultrasonic diagnosis modes have the following defects: the diagnosis efficiency is slow; doctors are in heavy workload and have high requirements on the experience level of the doctors, so the number of competent doctors is limited, the resources of the doctors are tense, and the cost is high.
Disclosure of Invention
The invention aims to provide an ultrasonic thyroid nodule intelligent evaluation method based on deep learning to assist a doctor in analyzing and judging the condition of a thyroid nodule so as to solve the existing technical problems.
In order to achieve the purpose, the invention adopts the technical scheme that: an ultrasonic thyroid nodule intelligent evaluation method based on deep learning is characterized by comprising the following steps:
s1, carrying out data annotation on the thyroid gland ultrasonic picture;
s2, performing data enhancement processing on the marked pictures;
s3, extracting the features of the data in the step S2 by using the resnet34 as a backbone network for extracting the features;
s4, generating a region suggestion frame by combining the suggestion network RPN with all the feature maps obtained in the step S3;
s5, obtaining the frame position of the target and the classification probability of the target;
s6, regarding the frame of the target with the classification probability P > M in the step S5 as an effective target, and intercepting the frame of the target as an input picture, wherein M is a threshold value;
s7, using ConvNet to extract the features of the input picture to obtain a feature map;
s8, taking the features obtained in the step S7 as shared features of different tasks, and respectively inputting different task classification functions;
s9, during training, a loss function L is set, and the classification function of each task is as follows:
L=α1L1+α2L2+α3L3+α4L4+α5L5
α1,α2…α5a weight corresponding to each task;
L1…L5=-∑tilnyi
yiis the value of output layer softmax, tiThe value is 0 or 1, the true target is 1, otherwise, the value is 0;
and S10, carrying out score statistics on different classifications, and judging the risk level according to the scores so as to give further diagnosis and treatment suggestions.
Further, in step S1, the data labeled information includes nodule position, composition, echo, shape, boundary and point-like hyperecho.
Further, in step S2, the classification data with less data is enhanced to achieve data balance.
Further, in step S7, ConvNet includes convolutional layers Conv, ReLU and pooling.
Further, in step S6, M is 0.8.
Further, in step S10, the formula for performing score statistics on different classifications is as follows
TR=CLASS1+CLASS2+CLASS3+CLASS4+CLASS5
Wherein, TR is thyroid risk grade score, and CLASS1 to CLASS5 are score values of components, echoes, shapes, boundaries and punctate hyperechoes respectively.
The invention has the beneficial technical effects that:
the invention adopts an automatic end-to-end deep learning method to automatically identify the thyroid ultrasonic picture, obtain all information of thyroid nodules, visually give final judgment and suggestions, assist doctors in analyzing and judging, reduce the workload of doctors, reduce the requirements on the doctors, improve the diagnosis efficiency and reduce the cost.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a flow chart of a method of the present invention;
fig. 2 is a schematic diagram of obtaining a region suggestion frame by using a suggestion network RPN according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a network structure of ConvNet according to an embodiment of the present invention.
Detailed Description
To further illustrate the various embodiments, the invention provides the accompanying drawings. The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the embodiments. Those skilled in the art will appreciate still other possible embodiments and advantages of the present invention with reference to these figures. Elements in the figures are not drawn to scale and like reference numerals are generally used to indicate like elements.
The invention will now be further described with reference to the accompanying drawings and detailed description.
As shown in fig. 1, an ultrasonic thyroid nodule intelligent assessment method based on deep learning is characterized by comprising the following steps:
and S1, carrying out data annotation on the thyroid gland ultrasonic picture.
Specifically, in this embodiment, the data annotation on the thyroid ultrasound image includes the following information: nodal location, composition, echo, shape, boundary, and point hyperecho information. Of course, in some embodiments, the information of the data label may be selected according to actual needs, for example, other information may be added on the basis of the above information, or only a part of the above information is selected.
And S2, performing data enhancement processing on the marked pictures.
Particularly, classified data with less data is enhanced to achieve data balance. The enhancement method may be operations such as adding noise to the picture, or performing rotational translation on the picture in a small range, or adjusting parameters such as color level and exposure level, and specifically refer to the prior art, which is not described in detail herein.
S3, feature extraction is performed on the data in step S2 using resnet34 as a backbone network for feature extraction.
The more the number of layers of the CNN (convolutional neural network), the richer the features that can be extracted, but simply increasing the number of layers of the convolutional can cause gradient dispersion or gradient explosion during training, and the resnet can effectively eliminate the problem of gradient dispersion or gradient explosion caused by the increase of the number of layers of the convolutional.
The resnet solves the problem of information loss and kernel loss in the traditional convolution in the information transmission process by changing the learning target, namely changing the complete learning output into the learning residual, protects the integrity of information by directly transmitting the input to the output in a detour mode, and reduces the learning difficulty by simplifying the learning target.
The specific principle of resnet34 can be referred to in particular in the prior art, which is not described in detail.
S4, generating a region suggestion box by using the suggested network RPN in combination with all the feature maps obtained in step S3, the generation process is shown in fig. 2, and specific principles can refer to the prior art, which can be easily implemented by those skilled in the art and will not be described in detail.
S5, the frame position of the target and the classification probability of the target are obtained.
The specific process can refer to the prior art, which can be easily realized by a person skilled in the art and is not described in detail.
And S6, regarding the frame of the target with the classification probability P > M in the step S5 as a valid target, namely the thyroid nodule target, and intercepting the frame of the target as an input picture, wherein M is a threshold value.
Specifically, in this embodiment, M is preferably >0.8, which improves the accuracy of the target, and of course, in other embodiments, the value of M may be selected according to actual needs.
S7, using ConvNet to perform feature extraction on the input picture, so as to obtain feature map (feature map).
Specifically, in this embodiment, the ConvNet includes convolutional layers Conv, ReLU and pooling, and the network structure is shown in fig. 3. The specific feature extraction process can refer to the prior art, which can be easily realized by a person skilled in the art, and is not described in detail.
And S8, respectively inputting the characteristics obtained in the step S7 into different task classification functions as shared characteristics of different tasks.
In this embodiment, the classification of each task is classified into 2 classes, 3 classes, or 4 classes as shown in the following table, but not limited thereto.
S9, during training, a loss function L is set, and the classification function of each task is as follows:
L=α1L1+α2L2+α3L3+α4L4+α5L5
α1,α2…α5the weight corresponding to each task can be specifically set according to the actual situation;
L1…L5=-∑tilnyi
yiis the value of output layer softmax, tiA value of 0 or 1, belonging to the true eyeLabeled 1, otherwise 0.
And S10, carrying out score statistics on different classifications, and judging the risk level according to the scores so as to give further diagnosis and treatment suggestions.
Specifically, in this embodiment, the formula for performing score statistics on different classifications is as follows:
TR=CLASS1+CLASS2+CLASS3+CLASS4+CLASS5
wherein, TR is thyroid risk grade score (risk grade can be obtained by score division according to the following table), CLASS1 to CLASS5 are score values of components, echoes, shapes, boundaries and punctate strong echoes respectively, different scores represent different categories, and specific category scores are shown in the following table, but not limited thereto.
Therefore, the final assessment and suggestion can be visually given to assist the doctor in analyzing and judging, so that the workload of the doctor is reduced, the requirement on the doctor is lowered, the diagnosis efficiency is improved, and the cost is lowered.
While the invention has been particularly shown and described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (6)
1. An ultrasonic thyroid nodule intelligent evaluation method based on deep learning is characterized by comprising the following steps:
s1, carrying out data annotation on the thyroid gland ultrasonic picture;
s2, performing data enhancement processing on the marked pictures;
s3, extracting the features of the data in the step S2 by using the resnet34 as a backbone network for extracting the features;
s4, generating a region suggestion frame by combining the suggestion network RPN with all the feature maps obtained in the step S3;
s5, obtaining the frame position of the target and the classification probability of the target;
s6, regarding the frame of the target with the classification probability P > M in the step S5 as an effective target, and intercepting the frame of the target as an input picture, wherein M is a threshold value;
s7, using ConvNet to extract the features of the input picture to obtain a feature map;
s8, taking the features obtained in the step S7 as shared features of different tasks, and respectively inputting different task classification functions;
s9, during training, a loss function L is set, and the classification function of each task is as follows:
L=α1L1+α2L2+α3L3+α4L4+α5L5
α1,α2…α5a weight corresponding to each task;
L1…L5=-∑tilnyi
yiis the value of output layer softmax, tiThe value is 0 or 1, the true target is 1, otherwise, the value is 0;
and S10, carrying out score statistics on different classifications, and judging the risk level according to the scores so as to give further diagnosis and treatment suggestions.
2. The deep learning based ultrasound thyroid nodule intelligent assessment method according to claim 1, wherein: in step S1, the information of the data annotation includes nodule position, composition, echo, shape, boundary, and point-like hyperecho.
3. The deep learning based ultrasound thyroid nodule intelligent assessment method according to claim 1, wherein: in step S2, the classification data with less data is enhanced to achieve data balance.
4. The deep learning based ultrasound thyroid nodule intelligent assessment method according to claim 1, wherein: in step S7, ConvNet includes convolutional layers Conv, ReLU and pooling.
5. The deep learning based ultrasound thyroid nodule intelligent assessment method according to claim 1, wherein: in step S6, M is 0.8.
6. The deep learning based ultrasound thyroid nodule intelligent assessment method according to claim 1, wherein: in step S10, the formula for performing score statistics on different classifications is as follows
TR=CLASS1+CLASS2+CLASS3+CLASS4+CLASS5
Wherein, TR is thyroid risk grade score, and CLASS1 to CLASS5 are score values of components, echoes, shapes, boundaries and punctate hyperechoes respectively.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010155463.0A CN111341443A (en) | 2020-03-09 | 2020-03-09 | Ultrasonic thyroid nodule intelligent evaluation method based on deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010155463.0A CN111341443A (en) | 2020-03-09 | 2020-03-09 | Ultrasonic thyroid nodule intelligent evaluation method based on deep learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111341443A true CN111341443A (en) | 2020-06-26 |
Family
ID=71187257
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010155463.0A Pending CN111341443A (en) | 2020-03-09 | 2020-03-09 | Ultrasonic thyroid nodule intelligent evaluation method based on deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111341443A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112419396A (en) * | 2020-12-03 | 2021-02-26 | 前线智能科技(南京)有限公司 | Thyroid ultrasonic video automatic analysis method and system |
CN112927808A (en) * | 2021-03-01 | 2021-06-08 | 北京小白世纪网络科技有限公司 | Thyroid ultrasound image-based nodule grading system and method |
CN113299391A (en) * | 2021-05-25 | 2021-08-24 | 李玉宏 | Risk assessment method for remote thyroid nodule ultrasonic image |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106295678A (en) * | 2016-07-27 | 2017-01-04 | 北京旷视科技有限公司 | Neural metwork training and construction method and device and object detection method and device |
CN107680678A (en) * | 2017-10-18 | 2018-02-09 | 北京航空航天大学 | Based on multiple dimensioned convolutional neural networks Thyroid ultrasound image tubercle auto-check system |
CN108734694A (en) * | 2018-04-09 | 2018-11-02 | 华南农业大学 | Thyroid tumors ultrasonoscopy automatic identifying method based on faster r-cnn |
CN109558896A (en) * | 2018-11-06 | 2019-04-02 | 中山大学附属第医院 | The disease intelligent analysis method and system with deep learning are learned based on ultrasound group |
CN109616195A (en) * | 2018-11-28 | 2019-04-12 | 武汉大学人民医院(湖北省人民医院) | The real-time assistant diagnosis system of mediastinum endoscopic ultrasonography image and method based on deep learning |
JP2020008992A (en) * | 2018-07-04 | 2020-01-16 | 株式会社日立製作所 | Data classification system, data classification method, and data classification device |
-
2020
- 2020-03-09 CN CN202010155463.0A patent/CN111341443A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106295678A (en) * | 2016-07-27 | 2017-01-04 | 北京旷视科技有限公司 | Neural metwork training and construction method and device and object detection method and device |
CN107680678A (en) * | 2017-10-18 | 2018-02-09 | 北京航空航天大学 | Based on multiple dimensioned convolutional neural networks Thyroid ultrasound image tubercle auto-check system |
CN108734694A (en) * | 2018-04-09 | 2018-11-02 | 华南农业大学 | Thyroid tumors ultrasonoscopy automatic identifying method based on faster r-cnn |
JP2020008992A (en) * | 2018-07-04 | 2020-01-16 | 株式会社日立製作所 | Data classification system, data classification method, and data classification device |
CN109558896A (en) * | 2018-11-06 | 2019-04-02 | 中山大学附属第医院 | The disease intelligent analysis method and system with deep learning are learned based on ultrasound group |
CN109616195A (en) * | 2018-11-28 | 2019-04-12 | 武汉大学人民医院(湖北省人民医院) | The real-time assistant diagnosis system of mediastinum endoscopic ultrasonography image and method based on deep learning |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112419396A (en) * | 2020-12-03 | 2021-02-26 | 前线智能科技(南京)有限公司 | Thyroid ultrasonic video automatic analysis method and system |
CN112419396B (en) * | 2020-12-03 | 2024-04-26 | 前线智能科技(南京)有限公司 | Automatic thyroid ultrasonic video analysis method and system |
CN112927808A (en) * | 2021-03-01 | 2021-06-08 | 北京小白世纪网络科技有限公司 | Thyroid ultrasound image-based nodule grading system and method |
CN113299391A (en) * | 2021-05-25 | 2021-08-24 | 李玉宏 | Risk assessment method for remote thyroid nodule ultrasonic image |
CN113299391B (en) * | 2021-05-25 | 2023-11-03 | 李玉宏 | Risk assessment method for remote thyroid nodule ultrasound image |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20200303062A1 (en) | Medical image aided diagnosis method and system combining image recognition and report editing | |
CN109447940B (en) | Convolutional neural network training method, ultrasonic image identification and positioning method and system | |
CN109325942B (en) | Fundus image structure segmentation method based on full convolution neural network | |
CN111341443A (en) | Ultrasonic thyroid nodule intelligent evaluation method based on deep learning | |
CN109858540B (en) | Medical image recognition system and method based on multi-mode fusion | |
CN108257135A (en) | The assistant diagnosis system of medical image features is understood based on deep learning method | |
CN110464380A (en) | A kind of method that the ultrasound cross-section image of the late pregnancy period fetus of centering carries out quality control | |
CN110490892A (en) | A kind of Thyroid ultrasound image tubercle automatic positioning recognition methods based on USFaster R-CNN | |
CN111986211A (en) | Deep learning-based ophthalmic ultrasonic automatic screening method and system | |
CN108765392B (en) | Digestive tract endoscope lesion detection and identification method based on sliding window | |
CN111767952B (en) | Interpretable lung nodule benign and malignant classification method | |
CN110047082A (en) | Pancreatic Neuroendocrine Tumors automatic division method and system based on deep learning | |
Tang et al. | Segnet-based gland segmentation from colon cancer histology images | |
CN112529894A (en) | Thyroid nodule diagnosis method based on deep learning network | |
CN108764342B (en) | Semantic segmentation method for optic discs and optic cups in fundus image | |
CN111430025B (en) | Disease diagnosis model training method based on medical image data augmentation | |
Zhao et al. | Application of an attention U-Net incorporating transfer learning for optic disc and cup segmentation | |
CN113223005A (en) | Thyroid nodule automatic segmentation and grading intelligent system | |
CN112270667B (en) | TI-RADS-based integrated deep learning multi-tag identification method | |
CN113012163A (en) | Retina blood vessel segmentation method, equipment and storage medium based on multi-scale attention network | |
CN114972362A (en) | Medical image automatic segmentation method and system based on RMAU-Net network | |
CN110399899B (en) | Cervical OCT image classification method based on capsule network | |
CN115471701A (en) | Lung adenocarcinoma histology subtype classification method based on deep learning and transfer learning | |
CN111833321A (en) | Window-adjusting optimization-enhanced intracranial hemorrhage detection model and construction method thereof | |
Wu et al. | Fatty liver diagnosis using deep learning in ultrasound image |
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: 20200626 |
|
RJ01 | Rejection of invention patent application after publication |