CN107203995A - Endoscopic images intelligent analysis method and system - Google Patents
Endoscopic images intelligent analysis method and system Download PDFInfo
- Publication number
- CN107203995A CN107203995A CN201710434090.9A CN201710434090A CN107203995A CN 107203995 A CN107203995 A CN 107203995A CN 201710434090 A CN201710434090 A CN 201710434090A CN 107203995 A CN107203995 A CN 107203995A
- Authority
- CN
- China
- Prior art keywords
- image
- focal zone
- focus characteristic
- information
- focus
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
- G06T7/0014—Biomedical image inspection using an image reference approach
-
- 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
- G06T2207/10068—Endoscopic 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
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Endoscopes (AREA)
Abstract
The application provides a kind of endoscopic images intelligent analysis method and system, and wherein method includes:Obtain image and the auxiliary information with the image association;Analyze the image and carry out key words sorting to there is the image of focal zone;Extract the focal zone of the image and contrasted with the focus characteristic database that corresponding image classification is marked, output is directed to the focus characteristic information of the focal zone;The lesion properties information of the focal zone is directed to described in receiving and is shown.The application is by obtaining image and focal zone to image and the focal zone chosen by interaction are handled, it can determine whether the genius morbi information included in image, complementary diagnostic information is provided for medical diagnosis, can solve the problem that and fail to pinpoint a disease in diagnosis disconnected, mistaken diagnosis caused by basic unit's physician specialty is lacked experience.
Description
Technical field
The application is related to endoscopic images analysis field, more particularly to a kind of endoscopic images intelligent analysis method and is
System.
Background technology
Current interaction diagnosis and therapy system only focuses on system in interactive mode, and health care worker is carrying out diagnosis and treatment work
When making, larger time cost need to be paid, fails really to facilitate doctor and patient.And diagnosis and therapy system depends on health care worker
Working experience, may result in Misdiagnosis under different medical experience levels.And interactive diagnosis and therapy system fails at present
Used with physician specialty experience, fail professional experiences really as a means of diagnosis and treatment, fail to play assisting in diagnosis and treatment
Effect.
In the prior art, current interactive system is concentrated mainly in information communication and information gathering, it is impossible to which realization has
The interactive scope image intellectual analysis of information exchange function.
The content of the invention
The application provides a kind of endoscopic images intelligent analysis method and system, to solve the deficiency in correlation technique.
According to the first aspect of the embodiment of the present application, there is provided a kind of endoscopic images intelligent analysis method, including following step
Suddenly:
Obtain image and the auxiliary information with the image association;
Analyze the image and carry out key words sorting to there is the image of focal zone;
Extract the focal zone of the image and contrasted with the focus characteristic database that corresponding image classification is marked, output is directed to
In the focus characteristic information of the focal zone;Wherein, the focal zone is to obtain the focal zone of the image automatically and/or pass through
The focal zone that interactive mode is chosen;
The lesion properties information of the focal zone is directed to described in receiving and is shown.
Further, the acquisition image and the auxiliary information with the image association, including:
The facility information for obtaining the image or video flowing of vision facilities transmission and being associated with the image or video flowing;
Described image or video flowing to acquisition are pre-processed.
Further, it is described to analyze the image and carry out key words sorting to there is the image of focal zone, including:
The image is analyzed, judges that the image has the image pair of focal zone with the presence or absence of focal zone and determination
Answer one or more in focus characteristic database.
Further, the focal zone for extracting the image and the focus characteristic database marked with corresponding image classification
Contrast, output is directed to the focus characteristic information of the focal zone, including:
The region that focus is there may be to the image carries out sliding window scanning, and the image that scanning is acquired is made in advance
For focal zone, and/or the focal zone that acquisition is chosen by interactive mode;
Extract the focal zone on the image and/or the focal zone chosen by interactive mode;
Analyze the focal zone extracted and contrasted with the focus characteristic database that corresponding image classification is marked, output is directed to
In the focus characteristic information of the focal zone.
Further, described endoscopic images intelligent analysis method also includes:
The focus characteristic for being directed to the lesion properties information of the focal zone described in receiving and being inputted by interactive mode
Information, if the focus characteristic information is not present in focus characteristic database, the focus characteristic information is stored in as newly-increased
Focus characteristic information focus characteristic self study database;
When newly-increased focus characteristic information in the self study database reaches predetermined threshold value, by the focus characteristic information
It is used as the training key element of focus characteristic self study database.
According to the first aspect of the embodiment of the present application there is provided a kind of endoscopic images intelligent analysis system, including:
Acquisition module, for obtaining image and auxiliary information with the image association;
Screening module, for analyzing the image and carrying out key words sorting to there is the image of focal zone;
Processing module, for the focus characteristic database for extracting the focal zone of the image and being marked with corresponding image classification
Contrast, output is directed to the focus characteristic information of the focal zone;Wherein, the focal zone is the disease for obtaining the image automatically
Stove area and/or the focal zone chosen by interactive mode;
Display module, described be directed to the lesion properties information of the focal zone and show for receiving.
Further, the acquisition module includes:
Subelement is obtained, for obtaining the image or video flowing of vision facilities transmission and being closed with the image or video flowing
The facility information of connection;
Pretreatment unit, is pre-processed for described image or video flowing to acquisition.
Further, the screening module includes:
Subelement is screened, for analyzing the image, the image is judged with the presence or absence of focal zone and determines there is disease
It is one or more in the image correspondence focus characteristic database in stove area.
Further, the processing module includes:
Unit is chosen, the region for there may be focus to the image carries out sliding window scanning, and scanning is obtained
Obtained image is pre- as focal zone, and/or obtains the focal zone chosen by interactive mode;
Extraction unit, for the focal zone for extracting focal zone on the image and/or being chosen by interactive mode;
Output unit is contrasted, the focal zone extracted for analyzing and the focus characteristic number marked with corresponding image classification
Contrasted according to storehouse, output is directed to the focus characteristic information of the focal zone.
Further, described endoscopic images intelligent analysis system also includes:
Self-learning module, described the lesion properties information of the focal zone is directed to and by interactive mode for receiving
The focus characteristic information of input, if the focus characteristic information is not present in focus characteristic database, the focus characteristic is believed
Breath deposit is used as newly-increased focus characteristic information focus characteristic self study database;
When newly-increased focus characteristic information in the self study database reaches predetermined threshold value, by the focus characteristic information
It is used as the training key element of focus characteristic self study database.
The technical scheme that embodiments herein is provided can include the following benefits:
The application can be sentenced by obtaining image and focal zone to image and the focal zone chosen by interaction are handled
The genius morbi information included in disconnected image, provides complementary diagnostic information for medical diagnosis, can solve the problem that because basic unit doctor is special
Industry fails to pinpoint a disease in diagnosis disconnected, mistaken diagnosis caused by lacking experience.
Brief description of the drawings
Fig. 1 is the flow chart of endoscopic images intelligent analysis method in the embodiment of the application one.
Fig. 2 is the flow chart of endoscopic images intelligent analysis method in the another embodiment of the application.
Fig. 3 is a kind of flow chart of endoscopic images intelligent analysis method in another embodiment of the application.
Fig. 4 is a kind of module map of endoscopic images intelligent analysis system in the embodiment of the application one.
Embodiment
Here exemplary embodiment will be illustrated in detail, its example is illustrated in the accompanying drawings.Following description is related to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represent same or analogous key element.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the application.On the contrary, they be only with it is such as appended
The example of the consistent apparatus and method of some aspects be described in detail in claims, the application.
It is the purpose only merely for description specific embodiment in term used in this application, and is not intended to be limiting the application.
" one kind ", " described " and "the" of singulative used in the application and appended claims are also intended to including majority
Form, unless context clearly shows that other implications.It is also understood that term "and/or" used herein refers to and wrapped
It may be combined containing one or more associated any or all of project listed.
Fig. 1 is the flow chart of endoscopic images intelligent analysis method in the embodiment of the application one.As shown in figure 1, in shown
Sight glass image intelligent analysis method comprises the following steps:
In a step 101, image and the auxiliary information with the image association are obtained.
Pending image is obtained, the image can be hospital PACS (Picture Archiving and
Communication Systems abbreviation, means image archiving and communication system) picture of equipment or mobile device clap
The image taken the photograph.And the auxiliary information of the image association can be the facility information or medical matters work in record data source
Make the other information of personnel's addition, the auxiliary information can as the classification image foundation;Certainly, classify the image
Methods are varied, and the application is not limited herein.
In a step 102, analyze the image and carry out key words sorting to there is the image of focal zone.
In this step, the image can be first analyzed with the presence or absence of focal zone, and focal zone there will be according to judged result
The image carry out key words sorting.The key words sorting is used for focus characteristic database alternatively with the image contrast.
In step 103, the focus characteristic database for extracting the focal zone of the image and being marked with corresponding image classification
Contrast, output is directed to the focus characteristic information of the focal zone;Wherein, the focal zone is the disease for obtaining the image automatically
Stove area and/or the focal zone chosen by interactive mode.
In this step, the automatic focal zone that obtains can provide tentative diagnosis comparative information for health care worker, and lead to
Crossing interactive mode selection focal zone further can diagnose comparative information by precision preliminary.Wherein choosing focal zone by interactive mode can
To be health care worker (can also be other operating personnel) by operating the focus of the image that display interface frame is selected
Area.Can have multiple with the focus characteristic database of the image contrast, for example, can set up cloud server, the cloud service
Device includes the database that a variety of neutral nets are constituted, a kind of database of pathology of every kind of or a variety of neutral net correspondences, the high in the clouds
The focus characteristic database of server storage can include image information, the focus characteristic information associated with image information etc., example
Such as, ulcer or tumor image and ulcer level, therapeutic scheme in stomach etc..By the focal zone and cloud server of the image
In database contrast, you can draw or judge the lesion information that the image is included.The step can be health care worker
There is provided auxiliary information, point out the current system acquisition of doctor the information arrived, it is understood that there may be genius morbi, solve because of basic unit doctor
Failed to pinpoint a disease in diagnosis caused by professional experiences are not enough, the possibility of mistaken diagnosis.
At step 104, it is directed to the lesion properties information of the focal zone described in receiving and shows.
In this step, lesion information that above-mentioned image includes can be received and shown on the display device, so as to medical matters
Staff consults, or is shown by other means, for example, print.
From above-described embodiment, the application is by obtaining image and the focal zone to image and the disease by interaction selection
Stove area is handled, and can determine whether the genius morbi information included in image, complementary diagnostic information is provided for medical diagnosis, can
Solution fails to pinpoint a disease in diagnosis disconnected, mistaken diagnosis caused by basic unit's physician specialty is lacked experience.
Fig. 2 is the flow chart of endoscopic images intelligent analysis method in the another embodiment of the application.Below in conjunction with Fig. 2 institutes
The flow chart stated is further described with the assistant analysis of endoscopic images to the application.As shown in Fig. 2 the endoscope shadow
As intelligent analysis method comprises the following steps.
In step 201, image and its related information are obtained.
The facility information for obtaining the image or video flowing of vision facilities transmission and being associated with the image or video flowing.Example
Such as, towards the importing picture of hospital's PACS equipment, the image that the video picture shooting assembly of facing mobile apparatus is obtained, or towards regarding
Picture interception component of frequency stream etc. acquires pending picture.The auxiliary information of the image association can be record data
The other information of facility information or the health care worker addition in source, such as patient symptom, the auxiliary information can be with
It is used as the foundation for the image of classifying;Certainly, the methods of the classification image are varied, and the application is not limited herein
System.
In step 202., pre-treatment image.
The pending image or video flowing that are obtained in step 101 are pre-processed.The pending image or video flowing
On the display device, operating personnel can be intercepted using the touch component of the display device to video flowing to scheme for the display that can be interacted
Picture, then carries out including the pretreatment such as rotation, adjustment of colour temperature aberration, to complete the local picture processing of low performance.Then pass through
The image department for crossing pretreatment is sent to cloud server or local service by way of compressing packing using remote communication module
Device carries out initial screening diagnostic result.Wherein, remote communication module includes ciphering unit, push unit and disconnection reconnecting unit;Plus
Close unit can realize compressed package encryption function, and push unit can realize long-range push function, and disconnection reconnecting unit can realize that height can
By the local information communicating requirement of property, it is possible to achieve the telecommunication demand of high security.Remote communication module also includes information
Receiving module, to receive the information that cloud server or home server are provided.
In step 203, initial screening diagnostic result.
In this step, the image is analyzed, the image is judged with the presence or absence of focal zone and determines there is focal zone
Image correspondence focus characteristic database in it is one or more.
Compressed package in above-mentioned steps can be decompressed, read corresponding facility information, the specific processing network (example of distribution
Such as the neutral net of a certain correspondence endoscopic images of cloud server), the preliminary screening for disease.In one embodiment,
The presence or absence of image classification of diseases can be carried out, the interactive information of operating personnel, such as lesion segmentation information, auxiliary diagnosis letter is received
Breath etc..Then the corresponding Neural Network Data storehouse of the image is determined according to lesion segmentation information, complementary diagnostic information etc., by institute
State image to be classified (such as endoscopic images and its corresponding classification), and judge currently to treat with the contrast of Neural Network Data storehouse
Whether diagnosis imaging there may be disease problems.The step is the assisting in diagnosis and treatment to disease, is provided just after diagnostic result is provided
Step, which is seen a doctor, to be instructed and diagnosis and treatment opinion, is available for operating personnel and patient.
In step 204, user mutual the selection result.
In this step, operating personnel may participate in man-machine interaction, and diagnostic result interaction is shown on the display device, if diagnosis
As a result to have disease and being approved by operating personnel, then step S207 is performed;If diagnostic result is has disease and operating personnel deny,
Then perform step S205;If diagnostic result is operating personnel thinks there is disease without disease, operating personnel complete focal zone frame
Choosing, then performs step S208, if without disease and operating personnel's accreditation, diagnosis is completed.
In step 205, diagnosis terminates, and shows result.
For example show certain gastric disease.
In step 206, subscriber frame selects focal zone.
I.e. the manual frame of user selects the focal zone on the image.
In step 207, lesion segmentation.
Cloud server is by being scanned mode to the image and Probability Area frame being carried out on the pending image
Choosing.For example, completing the selection of suspicious region using sliding window strategy, the region that sliding window is slipped over is used as pending district
Domain.Cloud server suspicious region is marked and choose to should image specific network.In this step, operating personnel
Man-machine interaction can equally be carried out.For example, saying prompting suspicious lesions region in interactive display, operating personnel can voluntarily increase can
Doubtful focal area can also delete suspicious lesions region.Wherein, newly-increased suspicious lesions region and original suspicious lesions region
Collectively constitute new suspicious region.After confirming through operating personnel, it will confirm that information package is compressed, into characteristic extraction step
S208。
In a step 208, feature extraction and contrast.
In this step, including to the image region that there may be focus carries out sliding window scanning, scanning every time
The image acquired will scan the image acquired in advance as focal zone as the willing focal zone that can exist, and/or obtain
Take the focal zone chosen by interactive mode.Then extract the focal zone on the image and/or chosen by interactive mode
The characteristic information of focal zone, described information feature extraction includes carrying out the spies such as sharpness of border degree, color, surface flatness, form
The extraction levied.The focal zone of ultimate analysis extraction is simultaneously contrasted with the focus characteristic database that corresponding image classification is marked, defeated
Go out to be directed to the focus characteristic information of the focal zone.Such as rubicundity and form in certain image-region swell, then the disease
Stove information is " redness ".
In one embodiment, image data lesion segmentation obtained selects the focus of segmentation as input, therefrom button
Region (region of frame choosing), the region of button choosing will be used as the pictorial information of input.The pictorial information is transmitted to cloud server,
Contrasted, after the genius morbi for obtaining the correspondence pictorial information, matched corresponding with the genius morbi with corresponding database
List of diseases and treatment method.For example, using stomach pictorial information as input pictorial information, according to the stomach pictorial information with
The image classification marks in corresponding database and searches corresponding genius morbi, then matches corresponding with the genius morbi
List of diseases and treatment method for display.
In step 209, display comparison result.
The application also provides another endoscopic images intelligent analysis method.Fig. 3 is in a kind of in the embodiment of the application one
The flow chart of sight glass image intelligent analysis method, this method its comprise the following steps:
In step 301, image and the auxiliary information with the image association are obtained.
In step 302, analyze the image and carry out key words sorting to there is the image of focal zone.
In step 303, the focus characteristic database for extracting the focal zone of the image and being marked with corresponding image classification
Contrast, output is directed to the focus characteristic information of the focal zone;Wherein, the focal zone is the disease for obtaining the image automatically
Stove area and/or the focal zone chosen by interactive mode.
In step 304, the lesion properties information of the focal zone is directed to described in receiving and defeated by interactive mode
The focus characteristic information entered, if the focus characteristic information is not present in focus characteristic database, by the focus characteristic information
Focus characteristic self study database is stored in as newly-increased focus characteristic information;Newly-increased focus in the self study database is special
Reference breath is when reaching predetermined threshold value, using the focus characteristic information as focus characteristic self study database training key element.
For example, if there is new focus characteristic, cloud server will record this focus characteristic, when new focus characteristic
A certain threshold value is reached, then as training key element, when training key element to reach a certain threshold value, cloud server is by re -training
Network, otherwise enters row information storage, and readjust and focus, pathology linked character, return using high in the clouds information storage module
New diagnostic data information result.
The present embodiment part unlike the embodiments above is that the present embodiment also has interactive information input function, real
Do not include the self-learning function in the case of all features in existing database, improve constantly information availability.
Other the application also corresponds to above method embodiment and provides a kind of endoscopic images intelligent analysis system.Fig. 4 is this
Apply for a kind of module map of endoscopic images intelligent analysis system in an embodiment, the system includes:
Acquisition module 401, for obtaining image and auxiliary information with the image association;
Screening module 402, for analyzing the image and carrying out key words sorting to there is the image of focal zone;
Processing module 403, for the focus characteristic number for extracting the focal zone of the image and being marked with corresponding image classification
Contrasted according to storehouse, output is directed to the focus characteristic information of the focal zone;Wherein, the focal zone is to obtain the image automatically
Focal zone and/or the focal zone chosen by interactive mode;
Display module 404, described be directed to the lesion properties information of the focal zone and show for receiving.
Optionally, the acquisition module 401 includes:
Subelement is obtained, for obtaining the image or video flowing of vision facilities transmission and being closed with the image or video flowing
The facility information of connection;
Pretreatment unit, is pre-processed for described image or video flowing to acquisition.
Optionally, the screening module 402 includes:
Subelement is screened, for analyzing the image, the image is judged with the presence or absence of focal zone and determines there is disease
It is one or more in the image correspondence focus characteristic database in stove area.
Optionally, the processing module 403 includes:
Unit is chosen, the region for there may be focus to the image carries out sliding window scanning, and scanning is obtained
Obtained image is pre- as focal zone, and/or obtains the focal zone chosen by interactive mode;
Extraction unit, for the focal zone for extracting focal zone on the image and/or being chosen by interactive mode;
Output unit is contrasted, the focal zone extracted for analyzing and the focus characteristic number marked with corresponding image classification
Contrasted according to storehouse, output is directed to the focus characteristic information of the focal zone.
Optionally, in addition to:
Self-learning module, described the lesion properties information of the focal zone is directed to and by interactive mode for receiving
The focus characteristic information of input, if the focus characteristic information is not present in focus characteristic database, the focus characteristic is believed
Breath is used as newly-increased focus characteristic information deposit focus characteristic self study database;
When newly-increased focus characteristic information in the self study database reaches predetermined threshold value, by the focus characteristic information
It is used as the training key element of focus characteristic self study database.
For device embodiment, because it corresponds essentially to embodiment of the method, so related part is real referring to method
Apply the part explanation of example.Device embodiment described above is only schematical, wherein described be used as separating component
The unit of explanation can be or may not be physically separate, and the part shown as unit can be or can also
It is not physical location, you can with positioned at a place, or can also be distributed on multiple NEs.Can be according to reality
Selection some or all of module therein is needed to realize the purpose of application scheme.Those of ordinary skill in the art are not paying
In the case of going out creative work, you can to understand and implement.
The preferred embodiment of the application is the foregoing is only, not to limit the application, all essences in the application
God is with principle, and any modification, equivalent substitution and improvements done etc. should be included within the scope of the application protection.
Claims (10)
1. a kind of endoscopic images intelligent analysis method, it is characterised in that comprise the following steps:
Obtain image and the auxiliary information with the image association;
Analyze the image and carry out key words sorting to there is the image of focal zone;
Extract the focal zone of the image and contrasted with the focus characteristic database that corresponding image classification is marked, output is directed to institute
State the focus characteristic information of focal zone;Wherein, the focal zone is to obtain the focal zone of the image automatically and/or by interaction
The focal zone that mode is chosen;
The lesion properties information of the focal zone is directed to described in receiving and is shown.
2. endoscopic images intelligent analysis method according to claim 1, it is characterised in that the acquisition image and with
The auxiliary information of the image association, including:
The facility information for obtaining the image or video flowing of vision facilities transmission and being associated with the image or video flowing;
Described image or video flowing to acquisition are pre-processed.
3. endoscopic images intelligent analysis method according to claim 1, it is characterised in that the analysis image is simultaneously
Key words sorting is carried out to there is the image of focal zone, including:
The image is analyzed, the image is judged with the presence or absence of focus and determines the corresponding focus of the image that there is focal zone
It is one or more in property data base.
4. endoscopic images intelligent analysis method according to claim 1, it is characterised in that the extraction image
Focus is simultaneously contrasted with the focus characteristic database that corresponding image classification is marked, and output is directed to the focus characteristic letter of the focal zone
Breath, including:
The region that focus is there may be to the image carries out sliding window scanning, and the image that scanning is acquired is used as disease in advance
Stove area, and/or obtain the focal zone chosen by interactive mode;
Extract the focal zone on the image and/or the focal zone chosen by interactive mode;
Analyze the focal zone extracted and contrasted with the focus characteristic database that corresponding image classification is marked, output is directed to institute
State the focus characteristic information of focal zone.
5. endoscopic images intelligent analysis method according to claim 1, it is characterised in that also include:
The focus characteristic information for being directed to the lesion properties information of the focal zone described in receiving and being inputted by interactive mode,
If the focus characteristic information is not present in focus characteristic database, the focus characteristic information is believed as newly-increased focus characteristic
Breath deposit focus characteristic self study database;
When newly-increased focus characteristic information in the self study database reaches predetermined threshold value, using the focus characteristic information as
The training key element of focus characteristic self study database.
6. a kind of endoscopic images intelligent analysis system, it is characterised in that including:
Acquisition module, for obtaining image and auxiliary information with the image association;
Screening module, for analyzing the image and carrying out key words sorting to there is the image of focal zone;
Processing module, for the focus characteristic database pair for extracting the focal zone of the image and being marked with corresponding image classification
Than output is directed to the focus characteristic information of the focal zone;Wherein, the focal zone is the focus for obtaining the image automatically
Area and/or the focal zone chosen by interactive mode;
Display module, described be directed to the lesion properties information of the focal zone and show for receiving.
7. endoscopic images intelligent analysis system according to claim 6, it is characterised in that the acquisition module includes:
Subelement is obtained, for obtaining the image or video flowing of vision facilities transmission and associating with the image or video flowing
Facility information;
Pretreatment unit, is pre-processed for described image or video flowing to acquisition.
8. endoscopic images intelligent analysis system according to claim 6, it is characterised in that the screening module includes:
Subelement is screened, for analyzing the image, the image is judged with the presence or absence of focal zone and determines there is focal zone
Image correspondence focus characteristic database in it is one or more.
9. endoscopic images intelligent analysis system according to claim 6, it is characterised in that the processing module includes:
Unit is chosen, the region for there may be focus to the image carries out sliding window scanning, and scanning is acquired
Image in advance be used as focal zone, and/or obtain by interactive mode selection focal zone;
Extraction unit, for the focal zone for extracting focal zone on the image and/or being chosen by interactive mode;
Output unit is contrasted, the focal zone extracted for analyzing and the focus characteristic database marked with corresponding image classification
Contrast, output is directed to the focus characteristic information of the focal zone.
10. endoscopic images intelligent analysis system according to claim 6, it is characterised in that also include:
Self-learning module, described be directed to the lesion properties information of the focal zone and inputted by interactive mode for receiving
Focus characteristic information, if the focus characteristic information is not present in focus characteristic database, by the focus characteristic information make
For newly-increased focus characteristic information deposit focus characteristic self study database;
When newly-increased focus characteristic information in the self study database reaches predetermined threshold value, using the focus characteristic information as
The training key element of focus characteristic self study database.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710434090.9A CN107203995A (en) | 2017-06-09 | 2017-06-09 | Endoscopic images intelligent analysis method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710434090.9A CN107203995A (en) | 2017-06-09 | 2017-06-09 | Endoscopic images intelligent analysis method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107203995A true CN107203995A (en) | 2017-09-26 |
Family
ID=59907386
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710434090.9A Pending CN107203995A (en) | 2017-06-09 | 2017-06-09 | Endoscopic images intelligent analysis method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107203995A (en) |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107563123A (en) * | 2017-09-27 | 2018-01-09 | 百度在线网络技术(北京)有限公司 | Method and apparatus for marking medical image |
CN107644419A (en) * | 2017-09-30 | 2018-01-30 | 百度在线网络技术(北京)有限公司 | Method and apparatus for analyzing medical image |
CN107967946A (en) * | 2017-12-21 | 2018-04-27 | 武汉大学 | Operating gastroscope real-time auxiliary system and method based on deep learning |
CN108682013A (en) * | 2018-05-30 | 2018-10-19 | 广州众健医疗科技有限公司 | A kind of gastroscope image intelligent processing unit |
CN108877915A (en) * | 2018-06-07 | 2018-11-23 | 合肥工业大学 | The intelligent edge calculations system of minimally invasive video processing |
CN108962355A (en) * | 2018-07-06 | 2018-12-07 | 腾讯科技(深圳)有限公司 | A kind of association display methods, device and service equipment, the storage medium of image |
CN109166120A (en) * | 2018-09-11 | 2019-01-08 | 百度在线网络技术(北京)有限公司 | For obtaining the method and device of information |
CN109222859A (en) * | 2018-08-03 | 2019-01-18 | 广州高通影像技术有限公司 | A kind of intelligent endoscope image system with AI auxiliary diagnosis function |
CN109446627A (en) * | 2018-10-22 | 2019-03-08 | 青岛美迪康数字工程有限公司 | Endoscopic images recognition methods and device |
CN109637652A (en) * | 2018-11-21 | 2019-04-16 | 北京全域医疗技术有限公司 | A kind of remote collaboration hook Target process and system based on cloud |
CN109767417A (en) * | 2017-11-10 | 2019-05-17 | 太豪生医股份有限公司 | Lesion detection device and its method |
CN109935305A (en) * | 2017-12-18 | 2019-06-25 | 天启慧眼(北京)信息技术有限公司 | Detect the methods, devices and systems of image |
CN111310851A (en) * | 2020-03-03 | 2020-06-19 | 四川大学华西第二医院 | Artificial intelligence ultrasonic auxiliary system and application thereof |
CN111430014A (en) * | 2020-03-31 | 2020-07-17 | 杭州依图医疗技术有限公司 | Display method, interaction method and storage medium of glandular medical image |
CN111476772A (en) * | 2020-04-03 | 2020-07-31 | 北京推想科技有限公司 | Medical image-based focus analysis method and device |
CN111539944A (en) * | 2020-04-28 | 2020-08-14 | 安徽科大讯飞医疗信息技术有限公司 | Lung focus statistical attribute acquisition method and device, electronic equipment and storage medium |
CN112117001A (en) * | 2020-09-08 | 2020-12-22 | 深圳市智慧健康产业发展有限公司 | General practitioner intelligence image auxiliary diagnosis system |
-
2017
- 2017-06-09 CN CN201710434090.9A patent/CN107203995A/en active Pending
Cited By (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107563123A (en) * | 2017-09-27 | 2018-01-09 | 百度在线网络技术(北京)有限公司 | Method and apparatus for marking medical image |
US10755411B2 (en) | 2017-09-27 | 2020-08-25 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method and apparatus for annotating medical image |
CN107644419A (en) * | 2017-09-30 | 2018-01-30 | 百度在线网络技术(北京)有限公司 | Method and apparatus for analyzing medical image |
CN109767417A (en) * | 2017-11-10 | 2019-05-17 | 太豪生医股份有限公司 | Lesion detection device and its method |
CN109767417B (en) * | 2017-11-10 | 2023-11-03 | 太豪生医股份有限公司 | Focus detection device and method thereof |
CN109935305A (en) * | 2017-12-18 | 2019-06-25 | 天启慧眼(北京)信息技术有限公司 | Detect the methods, devices and systems of image |
CN107967946A (en) * | 2017-12-21 | 2018-04-27 | 武汉大学 | Operating gastroscope real-time auxiliary system and method based on deep learning |
CN108682013A (en) * | 2018-05-30 | 2018-10-19 | 广州众健医疗科技有限公司 | A kind of gastroscope image intelligent processing unit |
CN108877915A (en) * | 2018-06-07 | 2018-11-23 | 合肥工业大学 | The intelligent edge calculations system of minimally invasive video processing |
CN108962355A (en) * | 2018-07-06 | 2018-12-07 | 腾讯科技(深圳)有限公司 | A kind of association display methods, device and service equipment, the storage medium of image |
CN109222859A (en) * | 2018-08-03 | 2019-01-18 | 广州高通影像技术有限公司 | A kind of intelligent endoscope image system with AI auxiliary diagnosis function |
CN109166120A (en) * | 2018-09-11 | 2019-01-08 | 百度在线网络技术(北京)有限公司 | For obtaining the method and device of information |
CN109446627A (en) * | 2018-10-22 | 2019-03-08 | 青岛美迪康数字工程有限公司 | Endoscopic images recognition methods and device |
CN109637652A (en) * | 2018-11-21 | 2019-04-16 | 北京全域医疗技术有限公司 | A kind of remote collaboration hook Target process and system based on cloud |
CN111310851A (en) * | 2020-03-03 | 2020-06-19 | 四川大学华西第二医院 | Artificial intelligence ultrasonic auxiliary system and application thereof |
CN111310851B (en) * | 2020-03-03 | 2023-04-28 | 四川大学华西第二医院 | Artificial intelligence ultrasonic auxiliary system and application thereof |
CN111430014A (en) * | 2020-03-31 | 2020-07-17 | 杭州依图医疗技术有限公司 | Display method, interaction method and storage medium of glandular medical image |
CN111430014B (en) * | 2020-03-31 | 2023-08-04 | 杭州依图医疗技术有限公司 | Glandular medical image display method, glandular medical image interaction method and storage medium |
CN111476772A (en) * | 2020-04-03 | 2020-07-31 | 北京推想科技有限公司 | Medical image-based focus analysis method and device |
CN111539944A (en) * | 2020-04-28 | 2020-08-14 | 安徽科大讯飞医疗信息技术有限公司 | Lung focus statistical attribute acquisition method and device, electronic equipment and storage medium |
CN111539944B (en) * | 2020-04-28 | 2024-04-09 | 讯飞医疗科技股份有限公司 | Method, device, electronic equipment and storage medium for acquiring statistical attribute of lung focus |
CN112117001A (en) * | 2020-09-08 | 2020-12-22 | 深圳市智慧健康产业发展有限公司 | General practitioner intelligence image auxiliary diagnosis system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107203995A (en) | Endoscopic images intelligent analysis method and system | |
Elazary et al. | Interesting objects are visually salient | |
CN101681512B (en) | Vein pattern management system, vein pattern registration device, vein pattern authentication device, vein pattern registration method, and vein pattern authentication method | |
CN101247760A (en) | Method for diagnosing disease from tongue image | |
Johnson et al. | Digital capture of fingerprints in a disaster victim identification setting: a review and case study | |
CN103975364B (en) | Selection of images for optical examination of the cervix | |
JP2009513205A (en) | Image processing system especially used for diagnostic images | |
CN105883511A (en) | Face recognition control method and system for elevators of intelligent residential districts | |
JPWO2019230302A1 (en) | Learning data collection device, training data collection method and program, learning system, trained model, and endoscopic image processing device | |
CN109166618A (en) | System for distribution of out-patient department and point examine method | |
US20070058845A1 (en) | Method for detecting abnormalities in medical screening | |
JPWO2019130924A1 (en) | Image processing equipment, endoscopic systems, image processing methods, and programs | |
CN109919912A (en) | A kind of quality evaluating method and device of medical image | |
CN110634578A (en) | Remote intelligent inquiry method | |
CN106295545A (en) | Recognition of face based on generalized information system | |
CN107977958A (en) | A kind of image diagnosing method and device | |
CN112614586B (en) | Remote disease intelligent diagnosis system based on medical images and block chains | |
CN109493340A (en) | Esophagus fundus ventricularis varication assistant diagnosis system and method under a kind of gastroscope | |
JP2011048438A (en) | Apparatus and system for processing pathological diagnosis | |
CN110070062A (en) | A kind of system and method for the recognition of face based on binocular active infrared | |
CN107680670A (en) | Medical image identification and diagnostic system and the method for a kind of no territory restriction | |
Tushabe | Computer forensics for cyberspace crimes | |
KR102284669B1 (en) | Medicine image analysis system for clinical trial participant, and method thereof | |
Uemura et al. | Novel, high‐definition 3‐D endoscopy system with real‐time compression communication system to aid diagnoses and treatment between hospitals in T hailand | |
CN115394406A (en) | Medical image artificial intelligence rapid diagnosis deployment method based on mobile terminal and mobile internet |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170926 |