CN110477846A - A kind of digestive endoscopy off-note real-time mark method and system - Google Patents
A kind of digestive endoscopy off-note real-time mark method and system Download PDFInfo
- Publication number
- CN110477846A CN110477846A CN201910783621.4A CN201910783621A CN110477846A CN 110477846 A CN110477846 A CN 110477846A CN 201910783621 A CN201910783621 A CN 201910783621A CN 110477846 A CN110477846 A CN 110477846A
- Authority
- CN
- China
- Prior art keywords
- module
- entropy
- threshold value
- detection zone
- image
- 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B1/00—Instruments for performing medical examinations of the interior of cavities or tubes of the body by visual or photographical inspection, e.g. endoscopes; Illuminating arrangements therefor
- A61B1/00002—Operational features of endoscopes
- A61B1/00004—Operational features of endoscopes characterised by electronic signal processing
- A61B1/00009—Operational features of endoscopes characterised by electronic signal processing of image signals during a use of endoscope
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B1/00—Instruments for performing medical examinations of the interior of cavities or tubes of the body by visual or photographical inspection, e.g. endoscopes; Illuminating arrangements therefor
- A61B1/00002—Operational features of endoscopes
- A61B1/0002—Operational features of endoscopes provided with data storages
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B1/00—Instruments for performing medical examinations of the interior of cavities or tubes of the body by visual or photographical inspection, e.g. endoscopes; Illuminating arrangements therefor
- A61B1/00002—Operational features of endoscopes
- A61B1/00043—Operational features of endoscopes provided with output arrangements
- A61B1/00045—Display arrangement
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B1/00—Instruments for performing medical examinations of the interior of cavities or tubes of the body by visual or photographical inspection, e.g. endoscopes; Illuminating arrangements therefor
- A61B1/273—Instruments for performing medical examinations of the interior of cavities or tubes of the body by visual or photographical inspection, e.g. endoscopes; Illuminating arrangements therefor for the upper alimentary canal, e.g. oesophagoscopes, gastroscopes
- A61B1/2736—Gastroscopes
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Surgery (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Pathology (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Biophysics (AREA)
- Physics & Mathematics (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Optics & Photonics (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Gastroenterology & Hepatology (AREA)
- Signal Processing (AREA)
- Endoscopes (AREA)
Abstract
The present invention relates to medical data digging technology fields, specially a kind of digestive endoscopy off-note real-time mark method and system, including computer system, the computer system includes: image zooming-out module, and described image extraction module is used to obtain the stomach Conventional white endoscopic video stream of input;Region division module, the region division module are divided into multiple net regions, and set the stomach wall in the endoscopic video as detection zone for that will carry out grid dividing to the endoscopic video stream;Zone marker module, the zone marker module is marked detection zone for nearest neighbor method, it avoids because of working strength and doctor's subjective judgement caused by the working time is made mistakes, reduce doctor's work load, the efficiency for improving medical diagnosis work can pass through the training of neural network by neuroid computing unit, gradually optimize entropy and threshold value, and then the confidence level of result can be greatly improved.
Description
Technical field
The present invention relates to medical data digging technology field, specially a kind of digestive endoscopy off-note real-time mark method
And system.
Background technique
China's gastric cancer newly sends out gastric cancer 40.5 ten thousand, 32.5 ten thousand dead every year, accounts for 42.6% He of global total amount respectively
45.0%, it reduces China's incidence gastric cancer rate and the death rate is public health problem urgently to be resolved.Clinical research shows gastric cancer
Prognosis is closely related with therapeutic effect.For suffering from the patient of advanced gastric carcinoma, even if receiving the gastric cancer based on surgical operation
Resection operation, the postoperative five-year survival rate of patient is still below 30%, and patient's postoperative life quality is low, brings to family and society
Greatly burden.If patient receives endoscopy and treatment in early gastric caacer in time, five-year survival rate is up to 90%, even
Radical treatment can be carried out to early carcinoma of stomach under scope.Therefore, early discovery, early diagnosis, early treatment EGC, to reduction gastric cancer
Incidence and mortality, save medical resources are of great significance.
Disease refers to the benign disease of stomach before the cancer of stomach, is the Major Risk Factors for causing gastric cancer comprising chronic to wither
Contracting gastritis, polyp of stomach, gastric ulcer, residual stomach and gastritis verrucosa etc..As main gastric precancerous lesion atrophic gastritis, cancer
Variability is 8.6~13.8%, and China is 1.2~7.1%.Existing result of study shows periodically to supervise gastric precancerous lesion
It surveys, can make the recall rate of early carcinoma of stomach is more than 50%.And gastric ulcer has the canceration rate of 1-2%.Therefore to the cancer for having gastric cancer risk
Preceding Disease should carry out facilitating cost-effective monitoring in early days, to be intervened, reduce the generation of gastric cancer.
Currently, the monitoring of gastric precancerous lesion canceration mainly uses ordinary optical endoscopic biopsy technique.In endoscopy,
The narrow problem in the scope visual field is usually made troubles to doctor: such as due to the limitation in the visual field, doctor must be repeatedly in inspection
Ensure that all lesions are found to fail to pinpoint a disease in diagnosis in order to avoid bringing in target organ inner wall surface moving lens.Therefore, develop one kind
Reliably, quickly the exception based on digestion endoscopic images in face of big data quantity of auxiliary doctor progress endoscopy especially is special
It is very necessary to levy real-time mark system, which can be used for that doctor is assisted to carry out early carcinomatous change screening.
Summary of the invention
In order to solve the problem above-mentioned, the present invention provides a kind of digestive endoscopy off-note real-time mark method and system.
The present invention solves its technical problem using following technical scheme to realize:
The present invention provides a kind of digestive endoscopy off-note real-time mark system, including computer system, the calculating
Machine system includes:
Image zooming-out module, described image extraction module are used to obtain the stomach Conventional white endoscopic video stream of input;
Region division module, the region division module are drawn for that will carry out grid dividing to the endoscopic video stream
It is divided into multiple net regions, and sets the stomach wall in the endoscopic video as detection zone;
Zone marker module, the zone marker module are marked detection zone for nearest neighbor method, will be adjacent
The detection zone of the video frame is associated;
Optical-flow Feature computing module, the Optical-flow Feature computing module is for calculating in the marked detection zone
Optical-flow Feature, rejects the region of dark, remaining part of detection zone carries out image enhancement processing;
Image processing module, the image that the processing module is used to will test region are converted to data, obtain entropy and threshold
Value;
Model training module, the model training module are used for through image processing module in disease before early stage stomach cancer
Sight glass figure and lesions position information carry out model training, obtain entropy and threshold value;
Abnormality detection module, the abnormality detection module are used at by endoscopic video described in image processing module
Reason, obtains the entropy and threshold value of stomach off-note;
Display module is marked, the label display module makes a mark location information in sequence of pictures, and will label
It is mapped in original input stomach Conventional white endoscopic video stream, and the described stomach scope video flowing is shown in real time
Show.
Preferably, the zone marker module includes the first computing unit, the second computing unit, third computing unit and pass
Receipts or other documents in duplicate member;Wherein,
First computing unit is used to calculate the zeroth order square and first moment of the detection zone;
Second computing unit, for the detection zone to be calculated according to the zeroth order square and the first moment
Regional center;
The third computing unit is used for according to the detection zone center calculation mahalanobis distance;
The associative cell is used for the shortest detection zone phase of the mahalanobis distance of the adjacent video frame
That answers is associated.
Preferably, described image processing module includes that weighting direction histogram unit, entropy computing unit and threshold value are chosen more
New unit;Wherein,
The weighting direction histogram unit is used to obtain weighting direction histogram according to the Optical-flow Feature;
The entropy computing unit is used to calculate the entropy of the weighting direction histogram;
The threshold value chooses updating unit and is used to choose detection threshold value, straight according to the detection threshold value and the weighting direction
The entropy of square figure detects whether abnormal behaviour, and updates the detection threshold value.
Preferably, described image processing module further includes based on neuroid computing unit, the neuroid meter
Calculate unit optimization method include:
1) it initializes, sets entropy WijAnd Wjt, given threshold OjAnd RtAssign the random value in (- 1,1);
2) one group of input sample and target sample P are randomly selectedkAnd TkIt is supplied to neural network;Wherein,
3) input sample P is usedk, entropy WijWith threshold value OjCalculate the input value S of middle layer neural unitj,Then S is usedj,Pass through
Transmission function calculates middle layer and exports Bj,, such as following formula:
4) B of middle layer is utilizedj, entropy WjtWith threshold value RtCalculate the output matrix L of output layer each unitt, then pass through biography
The response C of delivery function calculating output layer neural unitt, such as following formula:
5) target sample T is utilizedkWith the response C of output layer neural unittOutput layer unit generalization error dt is calculated,
Such as following formula:
6) entropy W is utilizedjt, middle layer BjThe generalized error of middle layer is calculated with the generalized error dt of output layerSuch as following formula:
7) B of generalized the error dt and middle layer of output layer are utilizedjTo correct entropy WjtWith threshold value Rt;
8) the generalized error of middle layer is utilizedWith target sample PkTo correct entropy WijWith threshold value Oj;
9) next learning sample is chosen automatically and be supplied to neural network, be restored to step 3);
10) input sample and target sample are first gone at random from sample again, is restored to step 3), until neural network is complete
Office's error is less than default minimum, illustrates network convergence, and training terminates at this time.
Preferably, in described image preprocessing module image enhancement processing include: image normalization, inactive pixels cut,
Image smoothing, image sharpening and image scaling.
The present invention also provides a kind of labeling methods of digestive endoscopy off-note real-time mark system, including following step
It is rapid:
1) disease endoscope figure before early stage stomach cancer and lesions position information input model training module are trained, are obtained
To entropy and threshold value;
2) subject's stomach Conventional white endoscopic video stream is obtained using endoscopic images system equipment;
3) Conventional white endoscopic video stream is subjected to grid dividing, is divided into multiple net regions, determine that endoscope regards
Stomach wall in frequency is detection zone;
4) detection zone is marked in nearest neighbor method, and the detection zone of the adjacent video frame is associated
4) calculate the Optical-flow Feature in the marked detection zone, reject the region of dark, detection zone its
Remaining part carries out image enhancement processing;
5) by the image input picture processing module after image enhancement processing, weighting direction is obtained according to the Optical-flow Feature
Histogram, then calculate it is described weighting direction histogram entropy, finally choose detection threshold value, according to the detection threshold value and it is described plus
The entropy of power direction histogram detects whether abnormal behaviour, and updates the detection threshold value;
6) entropy and threshold value is compared then to mark in effective image sequence, simultaneously when entropy exceeds the range of threshold value
Label is mapped in former gastroscope video flowing;
7) real-time display diagnoses labeled gastroscope video flowing over the display so that doctor observes confirmation.
Compared with prior art, the beneficial effects of the present invention are: a kind of digestive endoscopy off-note of the present invention is real
When tagging system, be applicable to that disease before stomach cancer is carried out effectively to detect the presence of suspicious lesions, find lesion classification, and to disease
Kitchen range body position is accurately positioned, and helps to find minimal disease, lesion is avoided to omit;Alleviation doctor can be assisted high-strength
Degree, prolonged diagosis work, avoids because of working strength and doctor's subjective judgement caused by the working time is made mistakes, and reduces doctor's work
Bear, improve the efficiency of medical diagnosis work, by neuroid computing unit, can by the training of neural network,
Gradually optimize entropy and threshold value, and then the confidence level of result can be greatly improved.
Detailed description of the invention
Fig. 1 is a kind of digestive endoscopy off-note real-time mark system structure diagram of the present invention;
Fig. 2 is a kind of digestive endoscopy off-note real-time mark method flow diagram of the present invention.
Specific embodiment
Below in conjunction with the attached drawing in embodiment of the present invention, the technical solution in embodiment of the present invention is carried out clear
Chu is fully described by, it is clear that described embodiment is only some embodiments of the invention, rather than whole implementation
Mode.Based on the embodiments of the present invention, those of ordinary skill in the art are obtained without making creative work
Every other embodiment, shall fall within the protection scope of the present invention.
The present invention provides a kind of technical solution referring to FIG. 1-2: a kind of digestive endoscopy off-note real-time mark system,
Including computer system, the computer system includes:
Image zooming-out module, described image extraction module are used to obtain the stomach Conventional white endoscopic video stream of input;
Region division module, the region division module are drawn for that will carry out grid dividing to the endoscopic video stream
It is divided into multiple net regions, and sets the stomach wall in the endoscopic video as detection zone;
Zone marker module, the zone marker module are marked detection zone for nearest neighbor method, will be adjacent
The detection zone of the video frame is associated;
Optical-flow Feature computing module, the Optical-flow Feature computing module is for calculating in the marked detection zone
Optical-flow Feature, rejects the region of dark, remaining part of detection zone carries out image enhancement processing;
Image processing module, the image that the processing module is used to will test region are converted to data, obtain entropy and threshold
Value;
Model training module, the model training module are used for through image processing module in disease before early stage stomach cancer
Sight glass figure and lesions position information carry out model training, obtain entropy and threshold value;
Abnormality detection module, the abnormality detection module are used at by endoscopic video described in image processing module
Reason, obtains the entropy and threshold value of stomach off-note;
Display module is marked, the label display module makes a mark location information in sequence of pictures, and will label
It is mapped in original input stomach Conventional white endoscopic video stream, and the described stomach scope video flowing is shown in real time
Show.
As one embodiment of the present invention, the zone marker module includes the first computing unit, the second calculating list
Member, third computing unit and associative cell;Wherein,
First computing unit is used to calculate the zeroth order square and first moment of the detection zone;
Second computing unit, for the detection zone to be calculated according to the zeroth order square and the first moment
Regional center;
The third computing unit is used for according to the detection zone center calculation mahalanobis distance;
The associative cell is used for the shortest detection zone phase of the mahalanobis distance of the adjacent video frame
That answers is associated.
As one embodiment of the present invention, described image processing module includes weighting direction histogram unit, entropy meter
It calculates unit and threshold value chooses updating unit;Wherein,
The weighting direction histogram unit is used to obtain weighting direction histogram according to the Optical-flow Feature;
The entropy computing unit is used to calculate the entropy of the weighting direction histogram;
The threshold value chooses updating unit and is used to choose detection threshold value, straight according to the detection threshold value and the weighting direction
The entropy of square figure detects whether abnormal behaviour, and updates the detection threshold value.
As one embodiment of the present invention, described image processing module further includes calculating list based on neuroid
The optimization method of member, the neuroid computing unit includes:
1) it initializes, sets entropy WijAnd Wjt, given threshold OjAnd RtAssign the random value in (- 1,1);
2) one group of input sample and target sample P are randomly selectedkAnd TkIt is supplied to neural network;Wherein,
3) input sample P is usedk, entropy WijWith threshold value OjCalculate the input value S of middle layer neural unitj,Then S is usedj,Pass through
Transmission function calculates middle layer and exports Bj,, such as following formula:
4) B of middle layer is utilizedj, entropy WjtWith threshold value RtCalculate the output matrix L of output layer each unitt, then pass through biography
The response C of delivery function calculating output layer neural unitt, such as following formula:
5) target sample T is utilizedkWith the response C of output layer neural unittOutput layer unit generalization error dt is calculated,
Such as following formula:
6) entropy W is utilizedjt, middle layer BjThe generalized error of middle layer is calculated with the generalized error dt of output layerSuch as following formula:
7) B of generalized the error dt and middle layer of output layer are utilizedjTo correct entropy WjtWith threshold value Rt;
8) the generalized error of middle layer is utilizedWith target sample PkTo correct entropy WijWith threshold value Oj;
9) next learning sample is chosen automatically and be supplied to neural network, be restored to step 3);
10) input sample and target sample are first gone at random from sample again, is restored to step 3), until neural network is complete
Office's error is less than default minimum, illustrates network convergence, and training terminates at this time, by the training of neural network, gradually optimizes entropy
Value and threshold value, can greatly improve the confidence level of result.
As one embodiment of the present invention, image enhancement processing includes: that image is returned in described image preprocessing module
One change, inactive pixels cutting, image smoothing, image sharpening and image scaling.
The present invention also provides a kind of technical solutions: a kind of label side of digestive endoscopy off-note real-time mark system
Method, comprising the following steps:
1) disease endoscope figure before early stage stomach cancer and lesions position information input model training module are trained, are obtained
To entropy and threshold value;
2) subject's stomach Conventional white endoscopic video stream is obtained using endoscopic images system equipment;
3) Conventional white endoscopic video stream is subjected to grid dividing, is divided into multiple net regions, determine that endoscope regards
Stomach wall in frequency is detection zone;
4) detection zone is marked in nearest neighbor method, and the detection zone of the adjacent video frame is closed
Connection;
5) calculate the Optical-flow Feature in the marked detection zone, reject the region of dark, detection zone its
Remaining part carries out image enhancement processing;
6) by the image input picture processing module after image enhancement processing, weighting direction is obtained according to the Optical-flow Feature
Histogram, then calculate it is described weighting direction histogram entropy, finally choose detection threshold value, according to the detection threshold value and it is described plus
The entropy of power direction histogram detects whether abnormal behaviour, and updates the detection threshold value;
7) entropy and threshold value is compared then to mark in effective image sequence, simultaneously when entropy exceeds the range of threshold value
Label is mapped in former gastroscope video flowing;
8) real-time display diagnoses labeled gastroscope video flowing over the display so that doctor observes confirmation.
Although hereinbefore invention has been described by reference to embodiment, the scope of the present invention is not being departed from
In the case where, various improvement can be carried out to it and can replace component therein with equivalent.Especially, as long as being not present
Structural conflict, the various features in presently disclosed embodiment can be combined with each other use by any way, In
The description for not carrying out exhaustive to the case where these combinations in this specification is examined merely for the sake of omission length with what is economized on resources
Consider.Therefore, the invention is not limited to specific embodiments disclosed herein, but the institute including falling within the scope of the appended claims
There is technical solution.
Claims (6)
1. a kind of digestive endoscopy off-note real-time mark system, including computer system, it is characterised in that: the department of computer science
System includes:
Image zooming-out module, described image extraction module are used to obtain the stomach Conventional white endoscopic video stream of input;
Region division module, the region division module are divided into for that will carry out grid dividing to the endoscopic video stream
Multiple net regions, and the stomach wall in the endoscopic video is set as detection zone;
Zone marker module, the zone marker module is marked detection zone for nearest neighbor method, described in adjacent
The detection zone of video frame is associated;
Optical-flow Feature computing module, the Optical-flow Feature computing module are used to calculate the light stream in the marked detection zone
Feature, rejects the region of dark, remaining part of detection zone carries out image enhancement processing;
Image processing module, the image that the processing module is used to will test region are converted to data, obtain entropy and threshold value;
Model training module, the model training module are used for through image processing module to disease endoscope before early stage stomach cancer
Figure and lesions position information carry out model training, obtain entropy and threshold value;
Abnormality detection module, the abnormality detection module are used to be handled by endoscopic video described in image processing module,
Obtain the entropy and threshold value of stomach off-note;
Display module is marked, the label display module makes a mark location information in sequence of pictures, and label is mapped
Real-time display is carried out into original input stomach Conventional white endoscopic video stream, and to the described stomach scope video flowing.
2. a kind of digestive endoscopy off-note real-time mark system according to claim 1, it is characterised in that: the region
Mark module includes the first computing unit, the second computing unit, third computing unit and associative cell;Wherein,
First computing unit is used to calculate the zeroth order square and first moment of the detection zone;
Second computing unit, for the region of the detection zone to be calculated according to the zeroth order square and the first moment
Center;
The third computing unit is used for according to the detection zone center calculation mahalanobis distance;
The associative cell is used for the shortest detection zone of the mahalanobis distance of the adjacent video frame is corresponding
It is associated.
3. a kind of digestive endoscopy off-note real-time mark system according to claim 1, it is characterised in that: described image
Processing module includes that weighting direction histogram unit, entropy computing unit and threshold value choose updating unit;Wherein,
The weighting direction histogram unit is used to obtain weighting direction histogram according to the Optical-flow Feature;
The entropy computing unit is used to calculate the entropy of the weighting direction histogram;
The threshold value chooses updating unit for choosing detection threshold value, according to the detection threshold value and the weighting direction histogram
Entropy detect whether abnormal behaviour, and update the detection threshold value.
4. a kind of digestive endoscopy off-note real-time mark system according to claim 3, it is characterised in that: described image
Processing module further includes based on neuroid computing unit, and the optimization method of the neuroid computing unit includes:
1) it initializes, sets entropy WijAnd Wjt, given threshold OjAnd RtAssign the random value in (- 1,1);
2) one group of input sample and target sample P are randomly selectedkAnd TkIt is supplied to neural network;Wherein,
3) input sample P is usedk, entropy WijWith threshold value OjCalculate the input value S of middle layer neural unitj, then use Sj, pass through biography
Delivery function calculates middle layer and exports Bj, such as following formula:
4) B of middle layer is utilizedj, entropy WjtWith threshold value RtCalculate the output matrix L of output layer each unitt, then pass through transmitting letter
Number calculates the response C of output layer neural unitt, such as following formula:
5) target sample T is utilizedkWith the response C of output layer neural unittOutput layer unit generalization error dt is calculated, it is as follows
Formula:
6) entropy W is utilizedjt, middle layer BjThe generalized error of middle layer is calculated with the generalized error dt of output layer
Such as following formula:
7) B of generalized the error dt and middle layer of output layer are utilizedjTo correct entropy WjtWith threshold value Rt;
8) the generalized error of middle layer is utilizedWith target sample PkTo correct entropy WijWith threshold value Oj;
9) next learning sample is chosen automatically and be supplied to neural network, be restored to step 3);
10) input sample and target sample are first gone at random from sample again, is restored to step 3), until the neural network overall situation is missed
Difference is less than default minimum, illustrates network convergence, and training terminates at this time.
5. a kind of digestive endoscopy off-note real-time mark system according to claim 1, it is characterised in that: described image
Image enhancement processing includes: image normalization, inactive pixels cutting, image smoothing, image sharpening and image in preprocessing module
Scaling.
6. a kind of label using the real-time mark system of digestive endoscopy off-note described in any one of claim 1-5
Method, which comprises the following steps:
1) disease endoscope figure before early stage stomach cancer and lesions position information input model training module are trained, obtain entropy
Value and threshold value;
2) subject's stomach Conventional white endoscopic video stream is obtained using endoscopic images system equipment;
3) Conventional white endoscopic video stream is subjected to grid dividing, is divided into multiple net regions, determines in endoscopic video
Stomach wall be detection zone;
4) detection zone is marked in nearest neighbor method, and the detection zone of the adjacent video frame is associated
4) calculate the Optical-flow Feature in the marked detection zone, reject the region of dark, detection zone remaining
Part carries out image enhancement processing;
5) by the image input picture processing module after image enhancement processing, weighting direction Histogram is obtained according to the Optical-flow Feature
Figure, then the entropy of the weighting direction histogram is calculated, detection threshold value is finally chosen, according to the detection threshold value and the weighting side
Abnormal behaviour is detected whether to the entropy of histogram, and updates the detection threshold value;
6) it compares entropy and threshold value then to mark in effective image sequence when entropy exceeds the range of threshold value, while will mark
Note is mapped in former gastroscope video flowing;
7) real-time display diagnoses labeled gastroscope video flowing over the display so that doctor observes confirmation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910783621.4A CN110477846A (en) | 2019-08-23 | 2019-08-23 | A kind of digestive endoscopy off-note real-time mark method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910783621.4A CN110477846A (en) | 2019-08-23 | 2019-08-23 | A kind of digestive endoscopy off-note real-time mark method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110477846A true CN110477846A (en) | 2019-11-22 |
Family
ID=68553272
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910783621.4A Pending CN110477846A (en) | 2019-08-23 | 2019-08-23 | A kind of digestive endoscopy off-note real-time mark method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110477846A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111932507A (en) * | 2020-07-31 | 2020-11-13 | 苏州慧维智能医疗科技有限公司 | Method for identifying lesion in real time based on digestive endoscopy |
CN112001915A (en) * | 2020-09-01 | 2020-11-27 | 山东省肿瘤防治研究院(山东省肿瘤医院) | Endoscope image processing method and system and readable storage medium |
CN112464936A (en) * | 2020-12-11 | 2021-03-09 | 合肥维信诺科技有限公司 | Positioning method and positioning device for abnormal display area of display module |
CN113160167A (en) * | 2021-04-16 | 2021-07-23 | 重庆飞唐网景科技有限公司 | Medical image data extraction working method through deep learning network model |
CN113596368A (en) * | 2021-07-23 | 2021-11-02 | 深圳市警威警用装备有限公司 | Information acquisition method based on law enforcement recorder and law enforcement recorder |
CN113935993A (en) * | 2021-12-15 | 2022-01-14 | 武汉楚精灵医疗科技有限公司 | Enteroscope image recognition system, terminal device, and storage medium |
CN116965765A (en) * | 2023-08-01 | 2023-10-31 | 西安交通大学医学院第二附属医院 | Early gastric cancer endoscope real-time auxiliary detection system based on target detection algorithm |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101126929A (en) * | 2007-09-05 | 2008-02-20 | 东北大学 | Continuous miner remote real-time failure forecast and diagnosis method and device |
CN104123544A (en) * | 2014-07-23 | 2014-10-29 | 通号通信信息集团有限公司 | Video analysis based abnormal behavior detection method and system |
CN104656620A (en) * | 2015-01-21 | 2015-05-27 | 清华大学 | Comprehensive evaluation system for remanufacturing of heavy-duty machine tool |
US20180253839A1 (en) * | 2015-09-10 | 2018-09-06 | Magentiq Eye Ltd. | A system and method for detection of suspicious tissue regions in an endoscopic procedure |
CN108852268A (en) * | 2018-04-23 | 2018-11-23 | 浙江大学 | A kind of digestive endoscopy image abnormal characteristic real-time mark system and method |
-
2019
- 2019-08-23 CN CN201910783621.4A patent/CN110477846A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101126929A (en) * | 2007-09-05 | 2008-02-20 | 东北大学 | Continuous miner remote real-time failure forecast and diagnosis method and device |
CN104123544A (en) * | 2014-07-23 | 2014-10-29 | 通号通信信息集团有限公司 | Video analysis based abnormal behavior detection method and system |
CN104656620A (en) * | 2015-01-21 | 2015-05-27 | 清华大学 | Comprehensive evaluation system for remanufacturing of heavy-duty machine tool |
US20180253839A1 (en) * | 2015-09-10 | 2018-09-06 | Magentiq Eye Ltd. | A system and method for detection of suspicious tissue regions in an endoscopic procedure |
CN108852268A (en) * | 2018-04-23 | 2018-11-23 | 浙江大学 | A kind of digestive endoscopy image abnormal characteristic real-time mark system and method |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111932507A (en) * | 2020-07-31 | 2020-11-13 | 苏州慧维智能医疗科技有限公司 | Method for identifying lesion in real time based on digestive endoscopy |
CN112001915A (en) * | 2020-09-01 | 2020-11-27 | 山东省肿瘤防治研究院(山东省肿瘤医院) | Endoscope image processing method and system and readable storage medium |
CN112464936A (en) * | 2020-12-11 | 2021-03-09 | 合肥维信诺科技有限公司 | Positioning method and positioning device for abnormal display area of display module |
CN113160167A (en) * | 2021-04-16 | 2021-07-23 | 重庆飞唐网景科技有限公司 | Medical image data extraction working method through deep learning network model |
CN113160167B (en) * | 2021-04-16 | 2022-01-14 | 深圳市铱硙医疗科技有限公司 | Medical image data extraction working method through deep learning network model |
CN113596368A (en) * | 2021-07-23 | 2021-11-02 | 深圳市警威警用装备有限公司 | Information acquisition method based on law enforcement recorder and law enforcement recorder |
CN113935993A (en) * | 2021-12-15 | 2022-01-14 | 武汉楚精灵医疗科技有限公司 | Enteroscope image recognition system, terminal device, and storage medium |
CN116965765A (en) * | 2023-08-01 | 2023-10-31 | 西安交通大学医学院第二附属医院 | Early gastric cancer endoscope real-time auxiliary detection system based on target detection algorithm |
CN116965765B (en) * | 2023-08-01 | 2024-03-08 | 西安交通大学医学院第二附属医院 | Early gastric cancer endoscope real-time auxiliary detection system based on target detection algorithm |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110477846A (en) | A kind of digestive endoscopy off-note real-time mark method and system | |
Chen et al. | GasHis-Transformer: A multi-scale visual transformer approach for gastric histopathological image detection | |
Wang et al. | Using deep convolutional neural networks for multi-classification of thyroid tumor by histopathology: a large-scale pilot study | |
WO2020215593A1 (en) | Artificial intelligence-based automatic evaluation method and system on quality check of gastrointestinal endoscopy | |
CN109166105A (en) | The malignancy of tumor risk stratification assistant diagnosis system of artificial intelligence medical image | |
Riegler et al. | Eir—efficient computer aided diagnosis framework for gastrointestinal endoscopies | |
CN110852396A (en) | Sample data processing method for cervical image | |
CN106529555A (en) | DR (Digital Radiography) slice lung contour extraction method based on fully convolutional network | |
CN111312405A (en) | Health examination gastric cancer screening, evaluating and managing system | |
Ma et al. | Developing and verifying automatic detection of active pulmonary tuberculosis from multi-slice spiral CT images based on deep learning | |
CN109785320A (en) | It is a kind of to be classified to mammograms and known method for distinguishing using modified AlexNet model | |
Patro et al. | SCovNet: A skip connection-based feature union deep learning technique with statistical approach analysis for the detection of COVID-19 | |
CN112071418B (en) | Gastric cancer peritoneal metastasis prediction system and method based on enhanced CT image histology | |
Wang et al. | Localizing and identifying intestinal metaplasia based on deep learning in oesophagoscope | |
CN110974179A (en) | Auxiliary diagnosis system for stomach precancer under electronic staining endoscope based on deep learning | |
Bejakovic et al. | Analysis of Crohn's disease lesions in capsule endoscopy images | |
CN111584071A (en) | Artificial intelligent auxiliary diagnosis model for primary malignant bone tumor around knee joint | |
Hossain et al. | Deeppoly: deep learning based polyps segmentation and classification for autonomous colonoscopy examination | |
CN112419246B (en) | Depth detection network for quantifying esophageal mucosa IPCLs blood vessel morphological distribution | |
Saraiva et al. | Artificial intelligence and high-resolution anoscopy: automatic identification of anal squamous cell carcinoma precursors using a convolutional neural network | |
Su et al. | Evaluation of deep learning methods for early gastric cancer detection using gastroscopic images | |
CN113990456A (en) | Deep learning-based graphical analysis and screening method and system for early cancers of digestive tract | |
Li et al. | Pixel-level diabetic retinopathy lesion detection using multi-scale convolutional neural network | |
Li et al. | Research of deep learning on gastric cancer diagnosis | |
Wang et al. | Multispectral image under tissue classification algorithm in screening of cervical cancer |
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: 20191122 |
|
RJ01 | Rejection of invention patent application after publication |