CN107230198B - Gastroscope image intelligent processing method and processing device - Google Patents
Gastroscope image intelligent processing method and processing device Download PDFInfo
- Publication number
- CN107230198B CN107230198B CN201710434430.8A CN201710434430A CN107230198B CN 107230198 B CN107230198 B CN 107230198B CN 201710434430 A CN201710434430 A CN 201710434430A CN 107230198 B CN107230198 B CN 107230198B
- Authority
- CN
- China
- Prior art keywords
- information
- image
- gastroscope
- gastroscope image
- level
- 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.)
- Active
Links
- 238000012545 processing Methods 0.000 title claims abstract description 31
- 238000003672 processing method Methods 0.000 title claims abstract description 16
- 238000004458 analytical method Methods 0.000 claims abstract description 30
- 238000000605 extraction Methods 0.000 claims description 21
- 230000003902 lesion Effects 0.000 claims description 15
- 239000000284 extract Substances 0.000 claims description 13
- 238000012216 screening Methods 0.000 claims description 11
- 230000011218 segmentation Effects 0.000 claims description 5
- 238000002372 labelling Methods 0.000 claims description 4
- 230000001419 dependent effect Effects 0.000 claims description 3
- 210000002784 stomach Anatomy 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims description 2
- 230000003287 optical effect Effects 0.000 claims description 2
- 230000000877 morphologic effect Effects 0.000 claims 2
- 238000003745 diagnosis Methods 0.000 abstract description 7
- 201000010099 disease Diseases 0.000 abstract description 7
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 abstract description 7
- 238000000034 method Methods 0.000 abstract description 7
- 230000007170 pathology Effects 0.000 abstract description 3
- 238000003384 imaging method Methods 0.000 description 5
- 230000002452 interceptive effect Effects 0.000 description 4
- 230000001575 pathological effect Effects 0.000 description 3
- 210000001519 tissue Anatomy 0.000 description 3
- 208000005718 Stomach Neoplasms Diseases 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 206010017758 gastric cancer Diseases 0.000 description 2
- 208000010749 gastric carcinoma Diseases 0.000 description 2
- 238000010827 pathological analysis Methods 0.000 description 2
- 201000011549 stomach cancer Diseases 0.000 description 2
- 201000000498 stomach carcinoma Diseases 0.000 description 2
- 206010028980 Neoplasm Diseases 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000001574 biopsy Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000002405 diagnostic procedure Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001861 endoscopic biopsy Methods 0.000 description 1
- 239000000686 essence Substances 0.000 description 1
- 210000001156 gastric mucosa Anatomy 0.000 description 1
- 206010020718 hyperplasia Diseases 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- 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
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30092—Stomach; Gastric
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Surgery (AREA)
- Optics & Photonics (AREA)
- Heart & Thoracic Surgery (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Radiology & Medical Imaging (AREA)
- Gastroenterology & Hepatology (AREA)
- Biomedical Technology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The application provides a kind of gastroscope image intelligent processing method and processing device, and wherein method includes：Dynamic access the first gastroscope image；Be partitioned on the first gastroscope image there are the cut zone of focus characteristic, marked on the first gastroscope image the corresponding cut zone level-one characteristic information and location information as the second gastroscope image, export the second gastroscope image；Receive the second gastroscope image, by in multiple second gastroscope images level-one characteristic information and location information be combined the zone position information that analysis forms secondary characteristics information and the corresponding secondary characteristics information, output is labeled with the third gastroscope image of the secondary characteristics information and the zone position information；It receives the third gastroscope image and shows.The application can help doctor to diagnose, and obtain pathology suggestion areas, greatly reduce fail to pinpoint a disease in diagnosis, the wrong probability examined.
Description
Technical field
This application involves gastroscope image processing field more particularly to a kind of gastroscope image intelligent processing method and processing devices.
Background technology
Currently, gastrocopy is still most direct, accurate, the reliable diagnostic method of gastric cancer.The precancerous lesion of gastric cancer is often
It is related to gastric mucosa extensive region, conventional endoscopic biopsy takes segmentation, random multidraw method, the focal cancerous issue of missing inspection
Possibility is very big, and the scope doctor that especially experience lacks then is easier " loseing depending on cancer " occur；Secondly, under the microscope for
The judgement of Atypical hyperplasia and early carcinoma of stomach needs a large amount of case accumulation, training, between different pathologists or same
One pathologist repeatedly diagnoses same group of gastroscopic biopsy sample, and all there is certain differences.
But traditional gastrocopy has a disadvantage that：1, early carcinoma of stomach diagnosis is low, and omission factor, misdiagnosis rate are high.2, existing
Gastroscope system only give doctor to provide image, and there is no carrying out primary dcreening operation and feature extraction to image and be labeled, and
Doctor has to face a large amount of medical image, and which increase the workload of doctor, efficiency is low and easily doctor is caused to go out
The case where existing " failing to pinpoint a disease in diagnosis ".Though prominent imaging 3, can be carried out to lesion, specific aim is had no for cancerous tissue, therefore to lesion nature
Judgement still by operation doctor experience.
Invention content
The application provides a kind of gastroscope image intelligent processing method and processing device, to solve deficiency in the related technology.
According to the embodiment of the present application in a first aspect, provide a kind of gastroscope image intelligent processing method, include the following steps：
Dynamic access the first gastroscope image；
It is partitioned on the first gastroscope image there are the cut zone of focus characteristic, in the first gastroscope image subscript
The level-one characteristic information and location information of the corresponding cut zone of note export the second gastroscope figure as the second gastroscope image
Picture；
The second gastroscope image is received, by the level-one characteristic information and location information in multiple second gastroscope images
It is combined the zone position information that analysis forms secondary characteristics information and the corresponding secondary characteristics information, output mark is
State the third gastroscope image of secondary characteristics information and the zone position information；
It receives the third gastroscope image and shows.
Further, described to be partitioned on the first gastroscope image there are the cut zone of focus characteristic, described
The level-one characteristic information for corresponding to the cut zone and location information are marked on one gastroscope image as the second gastroscope image, output
The second gastroscope image, including：
The first gastroscope image is scanned, to selecting and carrying out there are the regional frame of focus characteristic on the first gastroscope image
Label, extract level-one characteristic information in the cut zone of label and the cut zone in the gastroscope image
Location information；
Level-one characteristic information and the location information conduct of the corresponding cut zone are marked on the first gastroscope image
Second gastroscope image exports the second gastroscope image.
Further, described to receive the second gastroscope image, by the level-one feature in multiple second gastroscope images
Information and location information are combined analysis to form the regional location of secondary characteristics information and the corresponding secondary characteristics information
Information, output are labeled with the third gastroscope image of secondary characteristics information and zone position information, including：
It receives and carries the second gastroscope image, extract in multiple second gastroscopes and characteristic information and position letter
Breath is associated with multiple level-one characteristic informations, the associated multiple level-one feature letters of combinatory analysis according to the positional information
Breath forms secondary characteristics information, and determines the regional extent on the image of the secondary characteristics information, and output is labeled with described
The third gastroscope image of secondary characteristics information and the regional extent.
Further, described receive carries the second gastroscope image, extracts in multiple second gastroscopes and special
Reference ceases and location information, is associated with multiple level-one characteristic informations according to the positional information, combinatory analysis is associated multiple
The level-one characteristic information forms secondary characteristics information, and determines the regional extent on the image of the secondary characteristics information,
Output is labeled with the third gastroscope image of the secondary characteristics information and the regional extent, including：
It receives and carries the second gastroscope image, extract in multiple second gastroscopes and characteristic information and position letter
Breath is associated with multiple level-one characteristic informations, the associated multiple level-one feature letters of combinatory analysis according to the positional information
Cease the regional extent on the image for forming secondary characteristics information, determining the secondary characteristics information, the corresponding two level of retrieval
The feature description information of characteristic information, output are labeled with the secondary characteristics information, the feature description information and the region
The third gastroscope image of range.
Further, it is partitioned on the first gastroscope image that there are lesions after the first gastroscope of dynamic access image
Before the cut zone of feature, further include：
The gastroscope image of acquisition is subjected to initial screening, the gastroscope image is divided into the gastroscope image of no stove feature and ill
The gastroscope image of stove feature.
According to the second aspect of the embodiment of the present application, a kind of gastroscope image intelligent processing unit is provided, which is characterized in that packet
It includes：
Acquisition module is used for dynamic access the first gastroscope image；
First processing module, there are the cut zone of focus characteristic for being partitioned on the first gastroscope image, in institute
The level-one characteristic information that the corresponding cut zone is marked on the first gastroscope image and location information are stated as the second gastroscope image,
Export the second gastroscope image；
Second processing module, for receiving the second gastroscope image, by the level-one in multiple second gastroscope images
Characteristic information and location information are combined the region position that analysis forms secondary characteristics information and the corresponding secondary characteristics information
Confidence ceases, and output is labeled with the third gastroscope image of the secondary characteristics information and the zone position information；
Display module, for receiving the third gastroscope image and showing.
Further, the first processing module includes：
Pretreatment unit, for scanning the first gastroscope image, to there are focus characteristics on the first gastroscope image
Regional frame select and be marked, extract level-one characteristic information in the cut zone of label and the cut zone
Location information in the gastroscope image；
Output unit is marked, the level-one feature letter for marking the corresponding cut zone on the first gastroscope image
Breath and location information export the second gastroscope image as the second gastroscope image.
Further, the Second processing module includes：
First calculates output unit, carries the second gastroscope image for receiving, extracts in multiple second gastroscopes
And characteristic information and location information, be associated with multiple level-one characteristic informations according to the positional information, combinatory analysis is closed
Multiple level-one characteristic informations of connection form secondary characteristics information, and determine the area on the image of the secondary characteristics information
Domain range, output are labeled with the third gastroscope image of the secondary characteristics information and the regional extent.
Further, the Second processing module further includes：
Second calculates output unit, carries the second gastroscope image for receiving, extracts in multiple second gastroscopes
And characteristic information and location information, be associated with multiple level-one characteristic informations according to the positional information, combinatory analysis is closed
The region on the image that multiple level-one characteristic informations of connection form secondary characteristics information, determine the secondary characteristics information
Range, the feature description information of the corresponding secondary characteristics information of retrieval, output are labeled with the secondary characteristics information, the spy
Sign illustrates the third gastroscope image of information and the regional extent.
Further, the gastroscope image intelligent processing unit further includes：
Screening module, the gastroscope image for that will obtain carry out initial screening, the gastroscope image are divided into no stove feature
Gastroscope image and the gastroscope image for having focus characteristic.
The technical solution that embodiments herein provides can include the following benefits：
The application is split by the first gastroscope image to acquisition and is directed to cut zone and carried out feature extraction and mark
Note formed the second gastroscope image, then by multiple second gastroscope images carry out integrate obtain with can be for reference pathological characters believe
The third gastroscope image of breath, the third gastroscope image is given and is shown, doctor can be helped to diagnose, and obtains pathology and suggests area
Domain, greatly reduce fail to pinpoint a disease in diagnosis, the wrong probability examined.
Description of the drawings
Fig. 1 is a kind of flow chart of gastroscope image intelligent processing method in one embodiment of the application.
Fig. 2 is a kind of flow chart of gastroscope image intelligent processing method in another embodiment of the application.
Fig. 3 is a kind of module frame chart of gastroscope image intelligent processing unit in another embodiment of the application.
Specific implementation mode
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.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 consistent device 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, is not intended to be limiting the application.
It is also intended to including majority in the application and "an" of singulative used in the attached claims, " described " and "the"
Form, unless context clearly shows that other meanings.It is also understood that term "and/or" used herein refers to and wraps
Containing any or all combination of one or more associated list items purposes.
Fig. 1 is a kind of flow chart of gastroscope image intelligent processing method in one embodiment of the application.As shown in Figure 1, described
Gastroscope image intelligent processing method includes the following steps：
In step S101, dynamic access the first gastroscope image.
The first gastroscope image can obtain pending image by gastroscope in this step, be obtained by optical treatment
Obtain white light or NBI (NarrowBand Imaging scope Narrow-Band Imagings art) gastroscope image.The gastroscope image is to above-mentioned
The real-time dynamic access of gastroscopic apparatus, the dynamic image for obtaining presetting time interval can be specifically pre-seted by one.Institute
It states the operation that the first gastroscope image can be first passed through in advance including rotation, the adjustment of aberration colour temperature to be pre-processed, completes low performance
It is required that local image procossing.In addition, the first gastroscope image also can receive the information input of interactive display device.
In step s 102, it is partitioned on the first gastroscope image there are the cut zone of focus characteristic, described
The level-one characteristic information for corresponding to the cut zone and location information are marked on one gastroscope image as the second gastroscope image, output
The second gastroscope image.
In this step, receive the first gastroscope image, using algorithm of target detection to the first gastroscope image into
Row region division, including lesion separation and feature extraction.For example, by there may be diseases on the first gastroscope image
Stove characteristic area is scanned, on the first gastroscope image there may be the regions of focus characteristic to be split, and it is right
Cut zone carries out suspicious region label.Then, using the obtained image information data of cut zone that frame selects as inputting, from the
It is scratched in one gastroscope image and selects cut zone, recorded the position relationship between cut zone, extract the image of the cut zone
Including level-one characteristic information, including carry out the extraction of the features such as sharpness of border degree, color, surface flatness, form.Finally exist
Preliminary mark is carried out on the first gastroscope image, to obtain the second gastroscope image.Second gastroscope figure described in the step
As display can be interacted on the display device.
In step s 103, the second gastroscope image is received, by the level-one feature in multiple second gastroscope images
Information and location information are combined the zone position that analysis forms secondary characteristics information and the corresponding secondary characteristics information
Breath, output are labeled with the third gastroscope image of the secondary characteristics information and the zone position information.
In this step, the second gastroscope image may carry multiple level-one characteristic informations and location information, can pass through
Two or more level-one characteristic informations and location information are combined analysis, multiple level-one features by operation rule
New feature will can be formed by operation rule, this feature becomes secondary characteristics information.Such as level-one characteristic information is certain figure
It is swelled with form as field color is rubescent, " redness " focus characteristic is judged as according to specific operation rule, it should " redness " lesion
Feature is secondary characteristics information.And the zone position information of the corresponding secondary characteristics information can then pass through multiple level-ones
Location information present in characteristic information determines.In one embodiment, the zone position information, that is, multiple level-ones characteristic information
The set in the region of rubicundity and form protuberance in residing location sets on the image, such as certain image-region is exactly institute
State the zone position information of secondary characteristics information.
In the present embodiment, the extraction of secondary characteristics information mainly obtains two aspect information, one is obtaining secondary characteristics
Information region that may be present, this region depend on level-one characteristic information position that may be present, the second is it is special to obtain two level
The specific features of reference breath, the focus characteristic information that the combinatory analysis dependent on level-one characteristic information obtains.It is described obtaining
After secondary characteristics information and the zone position information, can by zone position information described in the secondary characteristics information labeling (
The region that image upper ledge is selected) on, form third gastroscope image, then export on interactive display device for doctor refering to.
In step S104, receives the third gastroscope image and show.
Receive the third gastroscope image and with the third gastroscope image associated lesion information (two level described above
Characteristic information and the zone position information).In practical application, the third gastroscope is observed on the display device by local
Image, and noted the image needed in the pathological tissue being locally extracted with reference to mark feature description on third gastroscope image and circle
Afterwards, linked groups cell extractions is carried out, the operation of doctor is facilitated, while improved doctor's pathological analysis and locally diagnosing
Efficiency.
By above-described embodiment it is found that the application is split by the first gastroscope image to acquisition and is directed to cut zone
Carry out feature extraction simultaneously mark to form the second gastroscope image, then by multiple second gastroscope images carry out integrate obtain with for
The third gastroscope image of the pathological characters information of reference, the third gastroscope image is given and is shown, doctor can be helped to examine
It is disconnected, obtain pathology suggestion areas, greatly reduce fail to pinpoint a disease in diagnosis, the wrong probability examined.In addition, the application utilizes the extraction side of focus characteristic
Formula can improve imaging precision, range, carry out prominent imaging to lesion, especially have specific aim to cancerous tissue, can be accurate
Cancerous issue is marked out to assist diagnosis.
Fig. 2 is a kind of flow chart of gastroscope image intelligent processing method in another embodiment of the application.As shown in Fig. 2, institute
Stating gastroscope image intelligent processing method includes：
In step s 201, Image Acquisition.
Such as the dynamic image of gastroscopic apparatus acquisition is obtained, a time interval can be pre-set, is obtained in the time interval
Interior dynamic image is not necessarily to manual intervention, automatic to obtain.NBI gastroscopes image is obtained as the first gastroscope image.In some embodiment party
The presetting pretreatment mode that can also follow up in formula pre-processes the NBI gastroscopes image, it may for example comprise image pair
The processing such as burnt, increase image definition.
In step S202, initial screening image.
In this step, the NBI gastroscopes image of acquisition is subjected to initial screening, according to presetting screening rule by institute
NBI gastroscope images are stated to be divided into the NBI gastroscopes image of no stove feature and there are the NBI gastroscope images of focus characteristic.In one embodiment,
The image for meeting default feature can be filtered out from a large amount of image and carries out preliminary judgement, and will confirm that focus characteristic
Image reaches interactive display device and shows, formula judgement can be interacted by doctor.For example, doctor to be determined with to the figure of focus characteristic
As notifying server that the image is further processed, doctor is judged that disease-free image notification server divides the image
To disease-free one kind, operated without next step.In this way, facilitating doctor to the access of the NBI gastroscopes image, doctor is improved
Working efficiency.
In step S203, level-one feature extraction and mark.
It in this step, can be by scanning the first gastroscope image, to there are lesion spies on the first gastroscope image
The regional frame of sign is selected and is marked, and extracts level-one characteristic information in the cut zone of label and the cut zone
Location information in the gastroscope image.The level-one feature of the corresponding cut zone is marked on the first gastroscope image
Information and location information export the second gastroscope image as the second gastroscope image.
For example, after receiving the NBI gastroscopes image that primary dcreening operation diagnostic module transmits, using algorithm of target detection to NBI stomaches
Mirror image carries out region division and carries out first class index (level-one characteristic information) extraction.Extraction process including lesion segmentation and
Feature extraction, lesion segmentation be will by the NBI gastroscopes image there may be the sector scanning of lesion and on figure
Probability Area frame choosing is carried out, and carries out suspicious region label；The characteristic extraction procedure of suspicious region is mainly completed in feature extraction, will
Divide obtained image information data as input, selects cut zone from having divided to scratch in image, carry out sharpness of border degree, face
The extraction of the features such as color, surface flatness, form, and preliminary mark is carried out on the image, form the second gastroscope image.At this
In step, the second gastroscope image can equally be interacted on the display device, so that doctor consults, doctor be facilitated to understand image.
In step S204, combinatory analysis feature.
The second gastroscope image is carried for example, receiving, the second gastroscope image includes multiple first class index, can be by two
Kind or two or more above-mentioned first class index are combined analysis, can pass through multiple judgements by first class index to two-level index and advise
Then, level-one feature calculation is become into new feature, this feature becomes two-level index.Certainly, the decision rule can also be from branch
It holds interactive display device update and inputs new decision rule.For example, first class index is that rubicundity and form are swelled, two level
Index then indicates " redness " that multiple first class index are formed.
In step S205, secondary characteristics extraction and mark.
In this step, in multiple second gastroscopes and characteristic information and location information are extracted, according to institute's rheme
The multiple level-one characteristic informations of information association are set, the associated multiple level-one characteristic informations of combinatory analysis form secondary characteristics
Information.Then and the regional extent on the image of the secondary characteristics information is determined, circle notes the secondary characteristics information on the diagram
Position range.
In step S206, the choosing of image-region frame and explanation.
In this step, it after extracting the secondary characteristics information of response, can also retrieve in the database described in corresponding to
The feature description information (such as red and swollen feature and treatment means) of secondary characteristics information.Finally in NBI gastroscope image labelings
There are the secondary characteristics information, the feature description information and the regional extent, finally using the NBI gastroscope images as third
Gastroscope image exports.
In step S207, image is shown.
Third gastroscope image is shown on the display device, is consulted for doctor.It is said based on zone circle note and secondary characteristics information
Bright image can help doctor to carry out respective organization cell extraction and pathological analysis, improve working efficiency and the diagnosis of doctor
Accuracy.
Fig. 3 is a kind of module frame chart of gastroscope image intelligent processing unit in another embodiment of the application.As shown in figure 3,
The gastroscope image intelligent processing unit, including：
Acquisition module 301 is used for dynamic access the first gastroscope image；
First processing module 302, for being partitioned on the first gastroscope image there are the cut zone of focus characteristic,
The level-one characteristic information for corresponding to the cut zone and location information are marked on the first gastroscope image as the second gastroscope figure
Picture exports the second gastroscope image；
Second processing module 303, for receiving the second gastroscope image, by one in multiple second gastroscope images
Grade characteristic information and location information are combined the region that analysis forms secondary characteristics information and the corresponding secondary characteristics information
Location information, output are labeled with the third gastroscope image of the secondary characteristics information and the zone position information；
Display module 304, for receiving the third gastroscope image and showing.
Further, the first processing module 302 includes：
Pretreatment unit, for scanning the first gastroscope image, to there are focus characteristics on the first gastroscope image
Regional frame select and be marked, extract level-one characteristic information in the cut zone of label and the cut zone
Location information in the gastroscope image；
Output unit is marked, the level-one feature letter for marking the corresponding cut zone on the first gastroscope image
Breath and location information export the second gastroscope image as the second gastroscope image.
Further, the Second processing module 303 includes：
First calculates output unit, carries the second gastroscope image for receiving, extracts in multiple second gastroscopes
And characteristic information and location information, be associated with multiple level-one characteristic informations according to the positional information, combinatory analysis is closed
Multiple level-one characteristic informations of connection form secondary characteristics information, and determine the area on the image of the secondary characteristics information
Domain range, output are labeled with the third gastroscope image of the secondary characteristics information and the regional extent.
Further, the Second processing module 303 further includes：
Second calculates output unit, carries the second gastroscope image for receiving, extracts in multiple second gastroscopes
And characteristic information and location information, be associated with multiple level-one characteristic informations according to the positional information, combinatory analysis is closed
The region on the image that multiple level-one characteristic informations of connection form secondary characteristics information, determine the secondary characteristics information
Range, the feature description information of the corresponding secondary characteristics information of retrieval, output are labeled with the secondary characteristics information, the spy
Sign illustrates the third gastroscope image of information and the regional extent.
Further, the gastroscope image intelligent processing unit further includes：
Screening module, the gastroscope image for that will obtain carry out initial screening, the gastroscope image are divided into no stove feature
Gastroscope image and the gastroscope image for having focus characteristic.
For device embodiments, since it corresponds essentially to embodiment of the method, so related place is referring to method reality
Apply the part explanation of example.The apparatus embodiments described above are merely exemplary, wherein described be used as separating component
The unit of explanation may or may not be physically separated, and the component shown as unit can be or can also
It is not physical unit, you can be located at a place, or may be distributed over multiple network units.It can be according to actual
It needs that some or all of module therein is selected 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 foregoing is merely the preferred embodiments of the application, not limiting the application, all essences in the application
With within principle, any modification, equivalent substitution, improvement and etc. done should be included within the scope of the application protection god.
Claims (8)
1. a kind of gastroscope image intelligent processing method, which is characterized in that include the following steps：
Dynamic access the first gastroscope image, the first gastroscope image is to obtain pending image by gastroscope, at optics
Reason acquires white light or NBI gastroscope images；
The first gastroscope image is scanned, is partitioned on the first gastroscope image that there are the cut zone of focus characteristic, in institute
The level-one characteristic information that the corresponding cut zone is marked on the first gastroscope image and location information are stated as the second gastroscope image,
Export the second gastroscope image；By to there may be focus characteristic regions to be scanned on the first gastroscope image,
On the first gastroscope image there may be the regions of focus characteristic to be split, and to cut zone carry out suspicious region mark
Note；Then, the image information data cut zone that frame selects obtained is scratched from the first gastroscope image as input and selects segmentation
Region, record cut zone between position relationship, the level-one characteristic information that the image for extracting the cut zone includes, including
Carry out the extraction of sharpness of border degree, color, surface flatness, morphological feature；Finally carried out just on the first gastroscope image
The mark of step, to obtain the second gastroscope image；
Receive the second gastroscope image, by multiple second gastroscope images level-one characteristic information and location information carry out
Combinatory analysis forms the zone position information of secondary characteristics information and the corresponding secondary characteristics information, and output is labeled with described two
The third gastroscope image of grade characteristic information and the zone position information；The second gastroscope image carries multiple level-one feature letters
Two or more level-one characteristic informations and location information, are combined by breath and location information by operation rule
Analysis, multiple level-one features will form new feature by operation rule, and this feature becomes secondary characteristics information；Secondary characteristics are believed
The extraction of breath obtains two aspect information, one is obtaining secondary characteristics information region that may be present, this region depends on one
Believe dependent on level-one feature the second is obtaining the specific features of secondary characteristics information grade characteristic information position that may be present
The focus characteristic information that the combinatory analysis of breath obtains；It, will after obtaining the secondary characteristics information and the zone position information
The secondary characteristics information labeling forms third gastroscope image on the zone position information；
It receives the third gastroscope image and shows.
2. gastroscope image intelligent processing method according to claim 1, which is characterized in that described to receive second gastroscope
Image, by multiple second gastroscope images level-one characteristic information and location information be combined analysis with formed two level spy
Reference ceases and the zone position information of the corresponding secondary characteristics information, output are labeled with secondary characteristics information and zone position
The third gastroscope image of breath, including：
The second gastroscope image is received, level-one characteristic information and location information in multiple second gastroscopes are extracted, according to
The location information is associated with multiple level-one characteristic informations, and the associated multiple level-one characteristic informations of combinatory analysis form two
Grade characteristic information, and determine the regional extent on the image of the secondary characteristics information, output is labeled with the secondary characteristics
The third gastroscope image of information and the regional extent.
3. gastroscope image intelligent processing method according to claim 2, which is characterized in that described to receive second gastroscope
Image extracts level-one characteristic information and location information in multiple second gastroscopes, is associated with according to the positional information multiple
The level-one characteristic information, the associated multiple level-one characteristic informations of combinatory analysis form secondary characteristics information, and determine institute
State the regional extent on the image of secondary characteristics information, output is labeled with the secondary characteristics information and the regional extent
Third gastroscope image, including：
The second gastroscope image is received, level-one characteristic information and location information in multiple second gastroscopes are extracted, according to
The location information is associated with multiple level-one characteristic informations, and the associated multiple level-one characteristic informations of combinatory analysis form two
Grade characteristic information, the regional extent on the image for determining the secondary characteristics information, the corresponding secondary characteristics information of retrieval
Feature description information, output is labeled with the of the secondary characteristics information, the feature description information and the regional extent
Three gastroscope images.
4. gastroscope image intelligent processing method according to claim 1, which is characterized in that the first gastroscope of the dynamic access
Before being partitioned on the first gastroscope image there are the cut zone of focus characteristic after image, further include：
The gastroscope image of acquisition is subjected to initial screening, the gastroscope image is divided into the gastroscope image of no stove feature and has lesion special
The gastroscope image of sign.
5. a kind of gastroscope image intelligent processing unit, which is characterized in that including：Acquisition module is used for the first gastroscope of dynamic access
Image, the first gastroscope image are to obtain pending image by gastroscope, and white light or NBI stomaches are acquired by optical treatment
Mirror image；
First processing module is partitioned on the first gastroscope image that there are lesion spies for scanning the first gastroscope image
The cut zone of sign marks the level-one characteristic information and location information of the corresponding cut zone on the first gastroscope image
As the second gastroscope image, the second gastroscope image is exported；By to there may be lesion spies on the first gastroscope image
Sign region be scanned, on the first gastroscope image there may be the regions of focus characteristic to be split, and to segmentation
Region carries out suspicious region label；Then, the image information data cut zone that frame selects obtained is as input, from first stomach
It is scratched in mirror image and selects cut zone, record the position relationship between cut zone, the image for extracting the cut zone includes
Level-one characteristic information, include carry out sharpness of border degree, color, surface flatness, morphological feature extraction；Finally described
Preliminary mark is carried out on one gastroscope image, to obtain the second gastroscope image；
Second processing module receives the second gastroscope image, by the level-one characteristic information in multiple second gastroscope images
It is combined the zone position information that analysis forms secondary characteristics information and the corresponding secondary characteristics information with location information, it is defeated
Go out to be labeled with the third gastroscope image of the secondary characteristics information and the zone position information；The second gastroscope image carries
Multiple level-one characteristic informations and location information, can by operation rule by two or more level-one characteristic informations and
Location information is combined analysis, and multiple level-one features will form new feature by operation rule, and it is special that this feature becomes two level
Reference ceases；The extraction of secondary characteristics information obtains two aspect information, one is secondary characteristics information region that may be present is obtained,
This region depends on level-one characteristic information position that may be present, the second is the specific features of secondary characteristics information are obtained,
The focus characteristic information that combinatory analysis dependent on level-one characteristic information obtains；Obtaining the secondary characteristics information and the area
After location information domain, by the secondary characteristics information labeling on the zone position information, third gastroscope image is formed；
Display module, for receiving the third gastroscope image and showing.
6. gastroscope image intelligent processing unit according to claim 5, which is characterized in that the Second processing module packet
It includes：
First calculates output unit, and for receiving the second gastroscope image, the level-one extracted in multiple second gastroscopes is special
Reference ceases and location information, is associated with multiple level-one characteristic informations according to the positional information, combinatory analysis is associated multiple
The level-one characteristic information forms secondary characteristics information, and determines the regional extent on the image of the secondary characteristics information,
Output is labeled with the third gastroscope image of the secondary characteristics information and the regional extent.
7. gastroscope image intelligent processing unit according to claim 6, which is characterized in that the Second processing module is also wrapped
It includes：
Second calculates output unit, and for receiving the second gastroscope image, the level-one extracted in multiple second gastroscopes is special
Reference ceases and location information, is associated with multiple level-one characteristic informations according to the positional information, combinatory analysis is associated multiple
The regional extent on the image that the level-one characteristic information forms secondary characteristics information, determines the secondary characteristics information, inspection
Rope corresponds to the feature description information of the secondary characteristics information, and output is labeled with the secondary characteristics information, the feature description
The third gastroscope image of information and the regional extent.
8. gastroscope image intelligent processing unit according to claim 5, which is characterized in that further include：
Screening module, the gastroscope image for that will obtain carry out initial screening, the gastroscope image are divided into the gastroscope of no stove feature
Image and the gastroscope image for having focus characteristic.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710434430.8A CN107230198B (en) | 2017-06-09 | 2017-06-09 | Gastroscope image intelligent processing method and processing device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710434430.8A CN107230198B (en) | 2017-06-09 | 2017-06-09 | Gastroscope image intelligent processing method and processing device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107230198A CN107230198A (en) | 2017-10-03 |
| CN107230198B true CN107230198B (en) | 2018-09-18 |
Family
ID=59934809
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710434430.8A Active CN107230198B (en) | 2017-06-09 | 2017-06-09 | Gastroscope image intelligent processing method and processing device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107230198B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107730489A (en) * | 2017-10-09 | 2018-02-23 | 杭州电子科技大学 | Wireless capsule endoscope small intestine disease variant computer assisted detection system and detection method |
| CN109461147B (en) * | 2018-10-26 | 2020-05-19 | 广州金域医学检验中心有限公司 | Pathological labeling method and device applied to FOV picture of mobile terminal |
| CN112263740A (en) * | 2020-10-09 | 2021-01-26 | 湘南学院附属医院 | Clinical multi-head reinforced gastric lavage device for digestive system department and control method |
| CN112991325B (en) * | 2021-04-14 | 2021-08-17 | 上海孚慈医疗科技有限公司 | Intelligent coding-based speckled red-emitting image acquisition and processing method and system |
| 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 |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100573581C (en) * | 2006-08-25 | 2009-12-23 | 西安理工大学 | Semi-automatic partition method of lung CT image focus |
| CN100562291C (en) * | 2006-11-08 | 2009-11-25 | 沈阳东软医疗系统有限公司 | A kind of at CT treatment of picture device, method and system |
| EA024855B1 (en) * | 2012-07-10 | 2016-10-31 | Закрытое Акционерное Общество "Импульс" | Method for obtaining subtraction angiographic image |
| CN104616296A (en) * | 2015-01-23 | 2015-05-13 | 上海联影医疗科技有限公司 | Method and device for improving quality of radiotherapy images and radiotherapy system |
| CN105894517B (en) * | 2016-04-22 | 2019-05-07 | 北京理工大学 | CT image liver segmentation method and system based on feature learning |
| CN106097335B (en) * | 2016-06-08 | 2019-01-25 | 安翰光电技术（武汉）有限公司 | Alimentary canal lesion image identification system and recognition methods |
-
2017
- 2017-06-09 CN CN201710434430.8A patent/CN107230198B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN107230198A (en) | 2017-10-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107230198B (en) | Gastroscope image intelligent processing method and processing device | |
| Ueyama et al. | Application of artificial intelligence using a convolutional neural network for diagnosis of early gastric cancer based on magnifying endoscopy with narrow‐band imaging | |
| Cai et al. | Using a deep learning system in endoscopy for screening of early esophageal squamous cell carcinoma (with video) | |
| Mori et al. | Artificial intelligence and upper gastrointestinal endoscopy: Current status and future perspective | |
| CN109871735B (en) | Image analysis method and device and manufacturing method for learning deep learning algorithm | |
| RU2765619C1 (en) | Computer classification of biological tissue | |
| Brown et al. | Chromoscopy versus conventional endoscopy for the detection of polyps in the colon and rectum | |
| JP7076698B2 (en) | Image analysis method, image analysis device, program, learned deep learning algorithm manufacturing method and learned deep learning algorithm | |
| CA2966555C (en) | Systems and methods for co-expression analysis in immunoscore computation | |
| CN111655116A (en) | Image diagnosis support device, data collection method, image diagnosis support method, and image diagnosis support program | |
| Young Park et al. | Automated image analysis of digital colposcopy for the detection of cervical neoplasia | |
| Singh et al. | Observer agreement in the assessment of narrowband imaging system surface patterns in Barrett’s esophagus: a multicenter study | |
| Muto et al. | Improving visualization techniques by narrow band imaging and magnification endoscopy | |
| US9649013B2 (en) | System and method for diagnosing and treating disease | |
| Renner et al. | Optical classification of neoplastic colorectal polyps–a computer-assisted approach (the COACH study) | |
| US20120327211A1 (en) | Diagnostic information distribution device and pathology diagnosis system | |
| JP5469070B2 (en) | Method and system using multiple wavelengths for processing biological specimens | |
| US20200134820A1 (en) | Tumor boundary reconstruction using hyperspectral imaging | |
| Csutak et al. | Agreement between image grading of conventional (45) and ultra wide-angle (200) digital images in the macula in the Reykjavik eye study | |
| CN103442628A (en) | Medical instrument for examining the cervix | |
| Waddingham et al. | The evolving role of endoscopy in the diagnosis of premalignant gastric lesions | |
| CN114372951A (en) | Nasopharyngeal carcinoma positioning and segmenting method and system based on image segmentation convolutional neural network | |
| Kimura-Tsuchiya et al. | Magnifying endoscopy with blue laser imaging improves the microstructure visualization in early gastric cancer: comparison of magnifying endoscopy with narrow‐band imaging | |
| Genta et al. | Pre‐neoplastic states of the gastric mucosa–a practical approach for the perplexed clinician | |
| Puig et al. | Optical diagnosis for colorectal polyps: a useful technique now or in the future? |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |