CN108695001A - A kind of cancer lesion horizon prediction auxiliary system and method based on deep learning - Google Patents
A kind of cancer lesion horizon prediction auxiliary system and method based on deep learning Download PDFInfo
- Publication number
- CN108695001A CN108695001A CN201810778123.6A CN201810778123A CN108695001A CN 108695001 A CN108695001 A CN 108695001A CN 201810778123 A CN201810778123 A CN 201810778123A CN 108695001 A CN108695001 A CN 108695001A
- Authority
- CN
- China
- Prior art keywords
- image
- client
- server
- cancer
- lesion
- 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
- 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
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/94—Hardware or software architectures specially adapted for image or video understanding
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Endoscopes (AREA)
Abstract
The cancer lesion horizon prediction auxiliary system and method, system that the invention discloses a kind of based on deep learning include client and server-side;Client is used to monitor the image of current endoscopic assistance acquisition and is transferred to server-side, receives and show the analysis result of server-side feedback;Server-side is used to, according to the image acquired from client, judge the corresponding position of image and genius loci immediately, analysis result is fed back to client.The present invention carries out Image Acquisition when endoscopic assistance first, and client, which is triggered, obtains acquired endoscopic image, and is uploaded to server-side;Server-side reception endoscopic image calls convolutional neural networks model to carry out the analysis that whether qualified endoscopic image, position judgement and genius loci identify as parameter;Client is called according to the analysis result of acquisition and indicates that the picture at each position and the label of genius loci are overlapped displaying.The present invention can make a definite diagnosis early-stage cancer, and canceration range is irised out, easy to use, have significant society and economic value.
Description
Technical field
The invention belongs to medical detection technologies, and in particular to a kind of cancer lesion horizon prediction based on deep learning
Auxiliary system and method.
Background technology
With the enhancing of people's health consciousness and gradually increasing for digestive endoscopy inspection, tumour, early stage under gastrointestinal mucosal
The discovery of cancer and precancerous lesion increases rapidly.Traditional surgery is since operation wound is big, patient's recovery is slow, matter of living after operation
Amount the reasons such as is also greatly lowered and is gradually replaced through endoscopic interventional treatment technology.The excision of cancerous lesion under scope is come
It says, determines that excision extension is very crucial.If excision extension is excessive, although cancer site can be cut off totally, to disease
The quality of life of people has certain influence;Excision extension is too small, cuts off totally extent of disease, can cause to recur, serious meeting
Jeopardize patient vitals.In fact, the precise positioning to disease still acquires a certain degree of difficulty, slighter canceration is visually difficult to.For
Gastric Diseases by Spraying, edge determination is more difficult when especially merging intestinal metaplasia, and operation doctor needs to observe after with magnifying gastroscope being amplified, in conjunction with dye
Color just can be with precise positioning lesion boundary according to the difference of anterior pituitary-adrenal cortex axis.
If can irising out diseased region in time to cancer lesions position judges under scope, prompting the position of canceration
And range, it is aided with pathology and the cytolgical examination of live body when necessary, to the further detailed inspection prompting part of auxiliary operation person
Position, then the target made a definite diagnosis early-stage cancer and precisely treated in time can be reached.
Invention content
In order to solve the above technical problem, the present invention provides a kind of, and the cancer lesion horizon prediction based on deep learning is auxiliary
Auxiliary system and method can iris out diseased region in time, prompt the position of canceration to cancer lesions position judges under scope
And range, it is aided with pathology and the cytolgical examination of live body when necessary, to the further detailed inspection prompting part of auxiliary operation person
Position reaches the target made a definite diagnosis early-stage cancer and precisely treated in time.
Technical solution is used by the system of the present invention:A kind of cancer lesion horizon prediction auxiliary based on deep learning
System, it is characterised in that:Including client and server-side;
The client, image for monitoring the acquisition of current endoscopic assistance are simultaneously transferred to the server-side, receive and aobvious
Show the analysis result of the server-side feedback;
The server-side, for according to the image acquired from client, judging that the corresponding position of image and position are special immediately
Sign, feeds back to client by analysis result.
Technical solution is used by the method for the present invention:A kind of cancer lesion horizon prediction side based on deep learning
Method, which is characterized in that include the following steps:
Step 1:When endoscopic assistance carries out Image Acquisition, client, which is triggered, obtains acquired endoscopic image, and uploads
To server-side;
Step 2:Server-side receives endoscopic image as parameter, calls whether convolutional neural networks model carries out endoscopic image
Qualified, position judges and the analysis of genius loci identification;
Step 3:Client according to the analysis result of acquisition, call the label of the picture and genius loci that indicate each position into
Row superposition displaying;
Step 4:When identifying canceration, enables " Class Activation map combining technology " and iris out extent of disease.
Beneficial effects of the present invention are:The position image that can be checked to scope through the invention is identified, when
It was found that there are when diseased region, system can prompt operator, and enable " Class Activation map combining technology ", the portion of real-time display canceration
Position and range, while the accurate image for extracting and storing cancer site, iris out lesion range, and endoscopic minimally-invasive is carried out for operator
Operation provides the range at excision position, improves the accuracy of excision extension, provides safeguard for patient health.
Description of the drawings
Fig. 1 is the system structure diagram of the embodiment of the present invention;
Fig. 2 is the method flow diagram of the embodiment of the present invention.
Specific implementation mode
Understand for the ease of those of ordinary skill in the art and implement the present invention, with reference to the accompanying drawings and embodiments to this hair
It is bright to be described in further detail, it should be understood that implementation example described herein is merely to illustrate and explain the present invention, not
For limiting the present invention.
It please add Fig. 1, a kind of cancer lesion horizon prediction auxiliary system based on deep learning provided by the invention, including visitor
Family end and server-side;
The client of the present embodiment, the endoscopic image for monitoring and being uploaded by network current endoscopic assistance acquisition, connects
Receive and show the analysis result of feedback.Each client includes communication module and image demonstration module;Wherein, communication module is used
Analysis result is obtained in transmiting a request to server-side, and from server-side, is implemented as http communication modes;Image demonstration module
For the analysis result according to acquisition, calls and indicate that the picture at each position and the label of genius loci are overlapped displaying.This reality
It applies in example, image demonstration module is whole including an oesophagus, stomach, duodenal bulb and drop portion, jejunum, ileum, colon, anal canal
The background schematic diagram of body;Schematic diagram for the PNG format for indicating each position;For indicating that there are lesion (i.e. genius locis)
Red picture.The red point diagram of the picture and lesion (i.e. genius loci) that indicate each position is called according to the information that server-side is beamed back
The position to indicate endoscopic technic inspected is covered on background schematic diagram for piece and there are the positions of lesion.
The server-side of the present embodiment, according to the endoscopic image acquired from client, judges immediately for using J2EE frameworks
The corresponding position of endoscopic image, genius loci and with the presence or absence of early carninomatosis stove and its lesion range, analysis result is fed back to
Client.Server-side includes sample database, convolutional neural networks model and web service module.
The sample database of the present embodiment is used to store the sample of typical endoscopic image, including qualified picture library, position library
With genius loci library, what is stored in qualified picture library is the endoscopic image that shooting clearly meets clarity requirement, in the library of position
What is stored is the endoscopic image of the object form progress position mark in pairing trrellis diagram piece, and what is stored in genius loci library is pairing
Endoscopic image in trrellis diagram piece carries out the endoscopic image of extent of disease mark.Usually a complete endoscopy report needs to wrap
Containing pars oralis pharyngis, oesophagus, stomach bottom, cardia (distant view), cardia (close shot), body of stomach rear wall, body of stomach size be curved, body of stomach antetheca, stomach angle, stomach
Sinus, pylorus, duodenal bulb, drop portion, jejunum, ileum, ileocecus, the colon ascendens, hepatic flexure of colon, transverse colon, splenic flexure, colon descendens, second
Shape colon, rectum, 24 positions of anal canal picture, if find there are lesion or suspicious position also need close to carry out more details
Shooting.Therefore, include required all sites in the present embodiment, in the position library, i.e.,:Pars oralis pharyngis, oesophagus, stomach
Bottom, cardia (distant view), cardia (close shot), body of stomach rear wall, body of stomach size be curved, body of stomach antetheca, stomach angle, antrum, pylorus, duodenum
Bulb, drop portion, jejunum, ileum, ileocecus, the colon ascendens, hepatic flexure of colon, transverse colon, splenic flexure, colon descendens, sigmoid colon, rectum, anal canal;
Identify the endoscopic image after concrete position, further recognition site feature whether there is lesion characteristics and export.Model is certainly
Dynamic position and the record for judging that lesion occurs.
The genius loci of the present embodiment include NBI cancers, NBI be normal, NBI non-cancer lesion (including polyp, inflammatory bowel disease,
It is rotten to the corn), white light cancer and white light are normal, white light non-cancer lesion (including polyp, inflammatory bowel disease, erosion), specially structure number
Group.Here, it is emphasized that mentioned in the present embodiment " pars oralis pharyngis, oesophagus, stomach bottom, cardia (distant view), cardia are (close
Scape), body of stomach rear wall, body of stomach size be curved, body of stomach antetheca, stomach angle, antrum, pylorus, duodenal bulb, drop portion, jejunum, ileum,
Ileocecus, the colon ascendens, hepatic flexure of colon, transverse colon, splenic flexure, colon descendens, sigmoid colon, rectum, anal canal ", " NBI cancers, NBI be normal, NBI
Non-cancer lesion (including polyp, inflammatory bowel disease, erosion), white light cancer and white light are normal, white light non-cancer lesion (including polyp, inflammation
Disease property enteropathy, erosion) " not diagnosis to disease, the feature being intended only as in picture, it can be understood as a parameter, and it is right
Their judgement and identification are the aspect ratios pair to picture.
The convolutional neural networks model of the present embodiment includes being trained according to qualified picture library, position library and genius loci library
Three models, be respectively used to whether endoscopic image qualified, position judges, genius loci identification and the determination of cancer context.
Model is Resnet50, is developed using Python, is packaged into after RESTful API (network interface of REST style) by it
He calls module.It is conventional that the training process of convolutional neural networks model, convolutional neural networks model, which are used for field of image recognition,
Technological means is no longer repeated herein.
The Web service module of the present embodiment is used to receive the request of client, using the endoscopic image received as parameter
It calls convolutional neural networks model to carry out the analysis that whether qualified endoscopic image, position judgement and genius loci identify successively, obtains
Client is fed back to analysis result.Client records the position of lesion and range and is shown in real time.
See Fig. 2, a kind of cancer lesion horizon prediction method based on deep learning provided by the invention, including following step
Suddenly:
Step 1:When endoscopic assistance carries out Image Acquisition, client, which is triggered, obtains the endoscopic image of acquisition, and is uploaded to
Server-side.
Step 2:Server-side receives endoscopic image as parameter, calls whether convolutional neural networks model carries out endoscopic image
Qualified, position judges and the analysis of genius loci identification;
First determine whether endoscopic image is qualified picture, it is unqualified that analysis result is exported if unqualified.It is specific next
Say, judge endoscopic image whether complete display, be capable of providing useful diagnostic message, continue to judge if qualified, if
Unqualified picture then skips other steps and directly exports result " unqualified ".
After endoscopic image is judged as qualified picture, the concrete position in the endoscopic image and output are identified.Position includes
Pars oralis pharyngis, oesophagus, stomach bottom, cardia (distant view), cardia (close shot), body of stomach rear wall, body of stomach size be curved, body of stomach antetheca, stomach angle, stomach
Sinus, pylorus, duodenal bulb, drop portion, jejunum, ileum, ileocecus, the colon ascendens, hepatic flexure of colon, transverse colon, splenic flexure, colon descendens, second
Shape colon, rectum, anal canal;Identify the endoscopic image after concrete position, further recognition site feature whether there is lesion
Feature simultaneously exports.Model automatic decision lesion occur position and record.
Genius loci includes that NBI cancers, NBI be normal, NBI non-cancer lesion (including polyp, inflammatory bowel disease, erosion), white light
Cancer and white light are normal, white light non-cancer lesion (including polyp, inflammatory bowel disease, erosion).
Step 3:Client receives and shows analysis result;
According to the analysis result of acquisition, calls and indicate that the picture at each position and the label of genius loci are overlapped displaying.
In the present embodiment, the analysis result that client is beamed back according to server-side call indicate each position (i.e. pars oralis pharyngis, oesophagus, stomach bottom,
Cardia (distant view), cardia (close shot), body of stomach rear wall, body of stomach size be curved, body of stomach antetheca, stomach angle, antrum, pylorus, duodenal bulb
Portion, drop portion, jejunum, ileum, ileocecus, the colon ascendens, hepatic flexure of colon, transverse colon, splenic flexure, colon descendens, sigmoid colon, rectum, anal canal)
Picture and the red point of lesion (i.e. the label of genius loci) picture cover on background schematic diagram to indicate endoscopic technic inspected
Position and position there are lesion.
Before convolutional neural networks final output layer, the execution overall situation is averagely converged and is used as in convolution Feature Mapping
The feature for being fully connected layer exported needed for generating.
Step 4:When identifying canceration, enables " Class Activation map combining technology " and iris out extent of disease;
There are check points for Client-Prompt there are when canceration, according to " Class Activation map combining technology " real-time display canceration
Range, operator further detailed inspection can prompt position, it may be necessary to endoscopic minimally-invasive lesion part cutting is carried out, with
Reach the target for making a definite diagnosis and treating early-stage cancer.
" Class Activation map combining technology " is that the weight of output layer is projected back in convolution Feature Mapping to identify the important of image
Region, the overall situation averagely collect the spatial averaging of the Feature Mapping of the output each unit of the last one convolutional layer, calculate last
The weighted sum of the characteristic pattern of a convolution figure layer is to obtain Class Activation figure.The color depth of Class Activation figure and the confidence level positive of prediction
It closes.
The present invention has following outstanding advantages:When carrying out endoscopy, if finding lesion, diseased region spy can be identified
Sign, irises out diseased region in time, and utilizes the position and range of " Class Activation map combining technology " prompt canceration in time, auxiliary when necessary
With the pathology of live body and cytolgical examination, to which the further detailed inspection of auxiliary operation person prompts position, made a definite diagnosis with to reach and
Treat the target of early-stage cancer.It is easy to use, there is significant society and economic value.One server can correspond to several
Client, each client correspond to an endoscopic assistance.
It should be understood that the part that this specification does not elaborate belongs to the prior art.
It should be understood that the above-mentioned description for preferred embodiment is more detailed, can not therefore be considered to this
The limitation of invention patent protection range, those skilled in the art under the inspiration of the present invention, are not departing from power of the present invention
Profit requires under protected ambit, can also make replacement or deformation, each fall within protection scope of the present invention, this hair
It is bright range is claimed to be determined by the appended claims.
Claims (10)
1. a kind of cancer lesion horizon prediction auxiliary system based on deep learning, it is characterised in that:Including client and service
End;
The client, image for monitoring the acquisition of current endoscopic assistance are simultaneously transferred to the server-side, receive and display institute
State the analysis result of server-side feedback;
The server-side will for according to the image acquired from client, judging the corresponding position of image and genius loci immediately
Analysis result feeds back to client.
2. the cancer lesion horizon prediction auxiliary system according to claim 1 based on deep learning, it is characterised in that:Institute
It includes sample database, convolutional neural networks model and web service module to state server-side;
The sample database is used to store the sample of endoscopic image, including qualified picture library, position library and genius loci library;Institute
State stored in qualified picture library be meet clarity requirement endoscopic image;What is stored in the position library is pairing trrellis diagram piece
In object form carry out position mark endoscopic image;What is stored in the genius loci library is the scope in pairing trrellis diagram piece
Image carries out the endoscopic image of extent of disease mark;
The convolutional neural networks model includes according to qualified picture library, trained three moulds in position library and genius loci library
Type is respectively used to whether qualified endoscopic image, position judgement and genius loci identification;
The Web service module is used to receive the request of client, using the endoscopic image received as parameter call convolution god
Carry out whether endoscopic image is qualified, position judges and the analysis of genius loci identification through network model, when detecting cancer site
When, the image of cancer site is accurately extracted and stored, lesion range, the position of real-time display canceration and range is irised out, will analyze
As a result client is fed back to.
3. the cancer lesion horizon prediction auxiliary system according to claim 2 based on deep learning, it is characterised in that:Institute
State all sites that can be checked comprising scope in the library of position, including pars oralis pharyngis, oesophagus, stomach bottom, distant view cardia, close shot cardia,
Body of stomach rear wall, body of stomach size be curved, body of stomach antetheca, stomach angle, antrum, pylorus, duodenal bulb, drop portion, jejunum, ileum, return it is blind
Portion, the colon ascendens, hepatic flexure of colon, transverse colon, splenic flexure, colon descendens, sigmoid colon, rectum, anal canal.
4. the cancer lesion horizon prediction auxiliary system according to claim 1 based on deep learning, it is characterised in that:Institute
It includes communication module and image demonstration module to state server-side;
The communication module obtains analysis result for transmiting a request to server-side, and from server-side;
Described image demonstration module is used for the analysis result according to acquisition, calls the mark of the picture and genius loci that indicate each position
Note is overlapped displaying.
5. a kind of cancer lesion horizon prediction method based on deep learning, which is characterized in that include the following steps:
Step 1:When endoscopic assistance carries out Image Acquisition, client, which is triggered, obtains acquired endoscopic image, and is uploaded to clothes
Business end;
Step 2:Server-side receives endoscopic image as parameter, and convolutional neural networks model is called to carry out whether endoscopic image closes
Lattice, position judge and the analysis of genius loci identification;
Step 3:Client is called according to the analysis result of acquisition and indicates that the picture at each position and the label of genius loci are folded
Add displaying.
6. the cancer lesion horizon prediction method according to claim 5 based on deep learning, which is characterized in that step 2
Specific implementation process be:First determine whether endoscopic image is qualified picture, it is not conform to that analysis result is exported if unqualified
Lattice;After endoscopic image is judged as qualified picture, identifies the concrete position and genius loci in the endoscopic image and export;Work as inspection
When measuring cancer site, enables " Class Activation map combining technology ", accurately extract and store the image of cancer site, iris out lesion model
It encloses, analysis result is fed back to client by the position of real-time display canceration and range.
7. the cancer lesion horizon prediction method according to claim 5 based on deep learning, it is characterised in that:Step 3
In, there are check points for Client-Prompt there are when canceration, according to the range of " Class Activation map combining technology " real-time display canceration,
The further detailed inspection of operator prompts position, it may be necessary to carry out endoscopic minimally-invasive lesion part cutting, be made a definite diagnosis with reaching
With the target for the treatment of early-stage cancer.
8. the cancer lesion horizon prediction method based on deep learning described according to claim 6 or 7, it is characterised in that:Institute
It is that the weight of output layer is projected back in convolution Feature Mapping to identify the important area of image, entirely to state " Class Activation map combining technology "
Office averagely collects the spatial averaging of the Feature Mapping of the output each unit of the last one convolutional layer, calculates the last one trellis diagram
The weighted sum of the characteristic pattern of layer is to obtain Class Activation figure;The color depth of Class Activation figure and the confidence level positive correlation of prediction.
9. the cancer lesion horizon prediction method based on deep learning according to claim 6-7 any one, feature
It is:Genius loci described in step 2 includes NBI cancers, NBI is normal, NBI non-cancer lesion, white light cancer and white light are normal, white light is non-
Carninomatosis stove;Wherein NBI non-cancer lesion includes polyp, inflammatory bowel disease, erosion, and white light non-cancer lesion includes polyp, inflammatory bowel
Disease, erosion.
10. the cancer lesion horizon prediction method based on deep learning according to claim 6-7 any one, feature
It is:In step 3, before convolutional neural networks final output layer, the execution overall situation averagely converges and will in convolution Feature Mapping
It is used as generating the required feature for being fully connected layer exported.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810778123.6A CN108695001A (en) | 2018-07-16 | 2018-07-16 | A kind of cancer lesion horizon prediction auxiliary system and method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810778123.6A CN108695001A (en) | 2018-07-16 | 2018-07-16 | A kind of cancer lesion horizon prediction auxiliary system and method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108695001A true CN108695001A (en) | 2018-10-23 |
Family
ID=63850711
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810778123.6A Pending CN108695001A (en) | 2018-07-16 | 2018-07-16 | A kind of cancer lesion horizon prediction auxiliary system and method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108695001A (en) |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109411092A (en) * | 2018-11-28 | 2019-03-01 | 武汉大学人民医院（湖北省人民医院） | Meibomian gland Laser Scanning Confocal Microscope intellectual analysis assessment system and method based on deep learning |
| CN109447987A (en) * | 2018-11-28 | 2019-03-08 | 武汉大学人民医院（湖北省人民医院） | Ulcerative colitis assistant diagnosis system and method under colonoscopy based on deep learning |
| CN109584229A (en) * | 2018-11-28 | 2019-04-05 | 武汉大学人民医院（湖北省人民医院） | A kind of real-time assistant diagnosis system of Endoscopic retrograde cholangio-pancreatiography art and method |
| CN109599166A (en) * | 2018-11-28 | 2019-04-09 | 武汉大学人民医院（湖北省人民医院） | Meibomian gland infrared image based on deep learning assists in identifying system and method |
| CN109598708A (en) * | 2018-11-28 | 2019-04-09 | 武汉大学人民医院（湖北省人民医院） | The system and method for cancer of pancreas is identified and positioned under a kind of endoscopic ultrasonography |
| 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 |
| CN109615633A (en) * | 2018-11-28 | 2019-04-12 | 武汉大学人民医院（湖北省人民医院） | Crohn disease assistant diagnosis system and method under a kind of colonoscopy based on deep learning |
| CN109820507A (en) * | 2019-03-25 | 2019-05-31 | 钟霁媛 | Disease aided diagnosis method and device |
| CN110197722A (en) * | 2019-05-31 | 2019-09-03 | 贵州精准健康数据有限公司 | AI-CPU system platform |
| CN110443797A (en) * | 2019-08-15 | 2019-11-12 | 重庆金山医疗技术研究院有限公司 | Lesion auxiliary judgement device |
| TWI681407B (en) * | 2018-12-07 | 2020-01-01 | 謝孟軒 | Computer-aided recognition system, its method and its computer program product thereof |
| CN111067473A (en) * | 2019-11-20 | 2020-04-28 | 于天成 | Gastric precancerous resection boundary marking system and method |
| WO2020215810A1 (en) * | 2019-04-25 | 2020-10-29 | 天津御锦人工智能医疗科技有限公司 | Image recognition-based narrowband image detection method for colonoscopy procedure |
| WO2020215804A1 (en) * | 2019-04-25 | 2020-10-29 | 天津御锦人工智能医疗科技有限公司 | Colonoscope feces and liquid feces detection method based on deep learning |
| WO2020215806A1 (en) * | 2019-04-25 | 2020-10-29 | 天津御锦人工智能医疗科技有限公司 | Video acquisition system for narrow cavitary space based on microprobe structure |
| CN111862090A (en) * | 2020-08-05 | 2020-10-30 | 武汉楚精灵医疗科技有限公司 | Method and system for esophageal cancer preoperative management based on artificial intelligence |
| WO2020224153A1 (en) * | 2019-05-07 | 2020-11-12 | 武汉楚精灵医疗科技有限公司 | Nbi image processing method based on deep learning and image enhancement, and application thereof |
| CN112201335A (en) * | 2020-07-23 | 2021-01-08 | 中国人民解放军总医院 | System and method for identifying structure in abdominal cavity under linear array ultrasonic endoscope |
| WO2021036863A1 (en) * | 2019-08-23 | 2021-03-04 | 王国华 | Deep learning-based diagnosis assistance system for early digestive tract cancer and examination apparatus |
| CN112464579A (en) * | 2021-02-02 | 2021-03-09 | 四川大学 | Identification modeling method for searching esophageal cancer lesion area based on evolutionary neural network structure |
| CN112614103A (en) * | 2020-12-18 | 2021-04-06 | 上海市胸科医院 | Bronchoscope image feature comparison marking system and method based on deep learning |
| CN113593707A (en) * | 2021-09-29 | 2021-11-02 | 武汉楚精灵医疗科技有限公司 | Stomach early cancer model training method and device, computer equipment and storage medium |
| CN113610847A (en) * | 2021-10-08 | 2021-11-05 | 武汉楚精灵医疗科技有限公司 | Method and system for evaluating stomach markers in white light mode |
| CN113990456A (en) * | 2021-11-09 | 2022-01-28 | 山东一泓电子科技有限公司 | Deep learning-based graphical analysis and screening method and system for early cancers of digestive tract |
| CN114283192A (en) * | 2021-12-10 | 2022-04-05 | 厦门影诺医疗科技有限公司 | Gastroscopy blind area monitoring method, system and application based on scene recognition |
| CN114565611A (en) * | 2022-04-28 | 2022-05-31 | 武汉大学 | Medical information acquisition method and related equipment |
| CN115578437A (en) * | 2022-12-01 | 2023-01-06 | 武汉楚精灵医疗科技有限公司 | Intestinal body focus depth data acquisition method and device, electronic equipment and storage medium |
| WO2023143014A1 (en) * | 2022-01-29 | 2023-08-03 | 王国华 | Endoscope-assisted inspection method and device based on artificial intelligence |
| CN116745861A (en) * | 2021-01-11 | 2023-09-12 | 翰林大学产学合作团 | Control method, device and program of lesion judgment system obtained through real-time image |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106909778A (en) * | 2017-02-09 | 2017-06-30 | 北京市计算中心 | A kind of Multimodal medical image recognition methods and device based on deep learning |
| CN107563999A (en) * | 2017-09-05 | 2018-01-09 | 华中科技大学 | A kind of chip defect recognition methods based on convolutional neural networks |
| CN107967946A (en) * | 2017-12-21 | 2018-04-27 | 武汉大学 | Operating gastroscope real-time auxiliary system and method based on deep learning |
-
2018
- 2018-07-16 CN CN201810778123.6A patent/CN108695001A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106909778A (en) * | 2017-02-09 | 2017-06-30 | 北京市计算中心 | A kind of Multimodal medical image recognition methods and device based on deep learning |
| CN107563999A (en) * | 2017-09-05 | 2018-01-09 | 华中科技大学 | A kind of chip defect recognition methods based on convolutional neural networks |
| CN107967946A (en) * | 2017-12-21 | 2018-04-27 | 武汉大学 | Operating gastroscope real-time auxiliary system and method based on deep learning |
Cited By (39)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109447987A (en) * | 2018-11-28 | 2019-03-08 | 武汉大学人民医院（湖北省人民医院） | Ulcerative colitis assistant diagnosis system and method under colonoscopy based on deep learning |
| CN109584229A (en) * | 2018-11-28 | 2019-04-05 | 武汉大学人民医院（湖北省人民医院） | A kind of real-time assistant diagnosis system of Endoscopic retrograde cholangio-pancreatiography art and method |
| CN109599166A (en) * | 2018-11-28 | 2019-04-09 | 武汉大学人民医院（湖北省人民医院） | Meibomian gland infrared image based on deep learning assists in identifying system and method |
| CN109598708A (en) * | 2018-11-28 | 2019-04-09 | 武汉大学人民医院（湖北省人民医院） | The system and method for cancer of pancreas is identified and positioned under a kind of endoscopic ultrasonography |
| 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 |
| CN109615633A (en) * | 2018-11-28 | 2019-04-12 | 武汉大学人民医院（湖北省人民医院） | Crohn disease assistant diagnosis system and method under a kind of colonoscopy based on deep learning |
| CN109411092A (en) * | 2018-11-28 | 2019-03-01 | 武汉大学人民医院（湖北省人民医院） | Meibomian gland Laser Scanning Confocal Microscope intellectual analysis assessment system and method based on deep learning |
| TWI681407B (en) * | 2018-12-07 | 2020-01-01 | 謝孟軒 | Computer-aided recognition system, its method and its computer program product thereof |
| CN109820507A (en) * | 2019-03-25 | 2019-05-31 | 钟霁媛 | Disease aided diagnosis method and device |
| CN109820507B (en) * | 2019-03-25 | 2024-03-08 | 钟霁媛 | Disease auxiliary diagnosis method and device |
| WO2020215810A1 (en) * | 2019-04-25 | 2020-10-29 | 天津御锦人工智能医疗科技有限公司 | Image recognition-based narrowband image detection method for colonoscopy procedure |
| CN111839445A (en) * | 2019-04-25 | 2020-10-30 | 天津御锦人工智能医疗科技有限公司 | Narrow-band imaging detection method in colonoscopy based on image recognition |
| WO2020215804A1 (en) * | 2019-04-25 | 2020-10-29 | 天津御锦人工智能医疗科技有限公司 | Colonoscope feces and liquid feces detection method based on deep learning |
| WO2020215806A1 (en) * | 2019-04-25 | 2020-10-29 | 天津御锦人工智能医疗科技有限公司 | Video acquisition system for narrow cavitary space based on microprobe structure |
| CN111839443A (en) * | 2019-04-25 | 2020-10-30 | 天津御锦人工智能医疗科技有限公司 | Cavity narrow space video acquisition system based on micro-probe structure |
| WO2020224153A1 (en) * | 2019-05-07 | 2020-11-12 | 武汉楚精灵医疗科技有限公司 | Nbi image processing method based on deep learning and image enhancement, and application thereof |
| CN110197722A (en) * | 2019-05-31 | 2019-09-03 | 贵州精准健康数据有限公司 | AI-CPU system platform |
| CN110443797A (en) * | 2019-08-15 | 2019-11-12 | 重庆金山医疗技术研究院有限公司 | Lesion auxiliary judgement device |
| WO2021036863A1 (en) * | 2019-08-23 | 2021-03-04 | 王国华 | Deep learning-based diagnosis assistance system for early digestive tract cancer and examination apparatus |
| CN111067473A (en) * | 2019-11-20 | 2020-04-28 | 于天成 | Gastric precancerous resection boundary marking system and method |
| CN112201335A (en) * | 2020-07-23 | 2021-01-08 | 中国人民解放军总医院 | System and method for identifying structure in abdominal cavity under linear array ultrasonic endoscope |
| CN111862090B (en) * | 2020-08-05 | 2023-10-10 | 武汉楚精灵医疗科技有限公司 | Method and system for esophageal cancer preoperative management based on artificial intelligence |
| CN111862090A (en) * | 2020-08-05 | 2020-10-30 | 武汉楚精灵医疗科技有限公司 | Method and system for esophageal cancer preoperative management based on artificial intelligence |
| CN112614103A (en) * | 2020-12-18 | 2021-04-06 | 上海市胸科医院 | Bronchoscope image feature comparison marking system and method based on deep learning |
| US11935239B2 (en) | 2021-01-11 | 2024-03-19 | Industry Academic Cooperation Foundation, Hallym University | Control method, apparatus and program for system for determining lesion obtained via real-time image |
| CN116745861A (en) * | 2021-01-11 | 2023-09-12 | 翰林大学产学合作团 | Control method, device and program of lesion judgment system obtained through real-time image |
| CN116745861B (en) * | 2021-01-11 | 2024-10-11 | 翰林大学产学合作团 | Control method, device and recording medium of lesion judgment system obtained through real-time image |
| CN112464579A (en) * | 2021-02-02 | 2021-03-09 | 四川大学 | Identification modeling method for searching esophageal cancer lesion area based on evolutionary neural network structure |
| CN113593707A (en) * | 2021-09-29 | 2021-11-02 | 武汉楚精灵医疗科技有限公司 | Stomach early cancer model training method and device, computer equipment and storage medium |
| CN113593707B (en) * | 2021-09-29 | 2021-12-14 | 武汉楚精灵医疗科技有限公司 | Stomach early cancer model training method and device, computer equipment and storage medium |
| CN113610847A (en) * | 2021-10-08 | 2021-11-05 | 武汉楚精灵医疗科技有限公司 | Method and system for evaluating stomach markers in white light mode |
| CN113610847B (en) * | 2021-10-08 | 2022-01-04 | 武汉楚精灵医疗科技有限公司 | Method and system for evaluating stomach markers in white light mode |
| CN113990456A (en) * | 2021-11-09 | 2022-01-28 | 山东一泓电子科技有限公司 | Deep learning-based graphical analysis and screening method and system for early cancers of digestive tract |
| CN114283192A (en) * | 2021-12-10 | 2022-04-05 | 厦门影诺医疗科技有限公司 | Gastroscopy blind area monitoring method, system and application based on scene recognition |
| WO2023143014A1 (en) * | 2022-01-29 | 2023-08-03 | 王国华 | Endoscope-assisted inspection method and device based on artificial intelligence |
| CN114565611B (en) * | 2022-04-28 | 2022-07-19 | 武汉大学 | Medical information acquisition method and related equipment |
| CN114565611A (en) * | 2022-04-28 | 2022-05-31 | 武汉大学 | Medical information acquisition method and related equipment |
| CN115578437B (en) * | 2022-12-01 | 2023-03-14 | 武汉楚精灵医疗科技有限公司 | Intestinal body focus depth data acquisition method and device, electronic equipment and storage medium |
| CN115578437A (en) * | 2022-12-01 | 2023-01-06 | 武汉楚精灵医疗科技有限公司 | Intestinal body focus depth data acquisition method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108695001A (en) | A kind of cancer lesion horizon prediction auxiliary system and method based on deep learning | |
| JP6657480B2 (en) | Image diagnosis support apparatus, operation method of image diagnosis support apparatus, and image diagnosis support program | |
| CN107967946B (en) | Gastroscope operation real-time auxiliary system and method based on deep learning | |
| CN110600122B (en) | Digestive tract image processing method and device and medical system | |
| JP6875709B2 (en) | A computer-readable recording medium that stores a disease diagnosis support method, a diagnosis support system, a diagnosis support program, and this diagnosis support program using endoscopic images of the digestive organs. | |
| CN108615037A (en) | Controllable capsule endoscopy operation real-time auxiliary system based on deep learning and operating method | |
| Fukuda et al. | Comparison of performances of artificial intelligence versus expert endoscopists for real-time assisted diagnosis of esophageal squamous cell carcinoma (with video) | |
| Mira et al. | Early diagnosis of oral cancer using image processing and artificial intelligence | |
| CN113164010A (en) | Method for supporting diagnosis of disease using endoscopic image of digestive organ, diagnosis support system, diagnosis support program, and computer-readable recording medium storing the diagnosis support program | |
| CN109584229A (en) | A kind of real-time assistant diagnosis system of Endoscopic retrograde cholangio-pancreatiography art and method | |
| KR102287364B1 (en) | System and method for detecting lesion in capsule endoscopic image using artificial neural network | |
| WO2021054477A2 (en) | Disease diagnostic support method using endoscopic image of digestive system, diagnostic support system, diagnostic support program, and computer-readable recording medium having said diagnostic support program stored therein | |
| WO2007119297A1 (en) | Image processing device for medical use and image processing method for medical use | |
| JPWO2020121906A1 (en) | Medical support system, medical support device and medical support method | |
| CN116745861B (en) | Control method, device and recording medium of lesion judgment system obtained through real-time image | |
| JP4749732B2 (en) | Medical image processing device | |
| US20200090548A1 (en) | Image processing apparatus, image processing method, and computer-readable recording medium | |
| US20230255467A1 (en) | Diagnostic imaging device, diagnostic imaging method, diagnostic imaging program, and learned model | |
| CN107658028A (en) | A kind of method for obtaining lesion data, identification lesion method and computer equipment | |
| CN111839428A (en) | Method for improving detection rate of colonoscope adenomatous polyps based on deep learning | |
| CN113112449A (en) | Intestinal cancer auxiliary diagnosis model construction and auxiliary diagnosis method and system | |
| KR102095730B1 (en) | Method for detecting lesion of large intestine disease based on deep learning | |
| JPH10197522A (en) | Pathological tissue diagnosis support device | |
| JPWO2020183936A1 (en) | Inspection equipment, inspection method and storage medium | |
| CN114391792B (en) | Tumor prediction method and device based on narrow-band imaging and imaging endoscope |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20190717 Address after: 430223 Room 001, Building D2, 10 Building, Phase III, Huacheng Avenue, Donghu New Technology Development Zone, Wuhan City, Hubei Province Applicant after: Wuhan Chujingling Medical Technology Co., Ltd. Address before: 430060 No. 99 Zhang Zhidong Road, Wuchang District, Hubei, Wuhan Applicant before: People's Hospital of Wuhan University (Hubei Provincial People's Hospital) |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181023 |
|
| RJ01 | Rejection of invention patent application after publication |