CN107274404A - A kind of pathological analysis system and method based on image - Google Patents
A kind of pathological analysis system and method based on image Download PDFInfo
- Publication number
- CN107274404A CN107274404A CN201710609478.8A CN201710609478A CN107274404A CN 107274404 A CN107274404 A CN 107274404A CN 201710609478 A CN201710609478 A CN 201710609478A CN 107274404 A CN107274404 A CN 107274404A
- Authority
- CN
- China
- Prior art keywords
- image
- pathological
- zonule
- lesion
- pathology
- 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.)
- Withdrawn
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/32—Determination of transform parameters for the alignment of images, i.e. image registration using correlation-based methods
-
- 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
- G06T2207/10016—Video; Image sequence
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical 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)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The present invention proposes a kind of pathological analysis system based on image, including：Image acquisition module, image processing module, picture recognition module and pathology storehouse；Image acquisition module is carried by toter and is delivered to the test position in human body, the focus image of captured in real-time lesions position；Image processing module receives the focus image that image acquisition module is shot, and image procossing is carried out to focus image；The pathological image for the corresponding focus that is stored with pathology storehouse and healthy image；The lesion image that picture recognition module exports image processing module carries out image recognition with the pathological image in pathology storehouse and healthy image, and system provides pathological diagnosis result according to image recognition result.The pathological analysis system based on image of the present invention, iconography is combined with medical treatment, effective distribution of the intelligent diagnostics and Expert Resources of focus is realized, and realizes the diagnosis of the difficult and complicated illness or malignant disorders of outlying district or medical under-developed area.
Description
Technical field
The present invention relates to medical imaging field, more particularly to a kind of pathological analysis system based on image further relates to one kind
Pathological analysis method based on image.
Background technology
China is populous nation, and has stepped into aging society, and the difficulty of getting medical service is the primary of puzzlement China medical industry
Problem.
The reason for the difficulty of getting medical service, is mainly the skewness of medical resource, and patient assessment is intended to large hospital and looks for expert, and
The quantity and energy of expert is limited, and can only go to a doctor a number of patient daily, cause patient that many needs go to a doctor can not and
When obtain medical treatment, delay the best opportunity made a definite diagnosis and treated.
Although China is directed to the medical hospital of the reform of medical field, the limited amount of expert, and expert always
It is fixed, it is impossible to fundamentally to realize fair allocat of the patient to medical resource.
As digitlization, internet and big data technology are developed rapidly, how the express of information for taking passage, by medical treatment
Resource realizes reasonable layout by way of information-based, is current China's medical field urgent problem to be solved.
The content of the invention
The present invention proposes a kind of pathological analysis system and method based on image, and iconography is combined with medical treatment, realizes
Effective distribution of the intelligent diagnostics and Expert Resources of focus.
The technical proposal of the invention is realized in this way：
A kind of pathological analysis system based on image, including image acquisition module, image processing module, picture recognition module
With pathology storehouse；
Image acquisition module is carried by toter and is delivered to the test position in human body, captured in real-time lesions position
Focus image；
Image processing module receives the focus image that image acquisition module is shot, and image procossing is carried out to focus image, first
First, the random two field picture in every millisecond of image data is extracted, is lesion image by focus video conversion；Then, to focus figure
As carrying out similarity screening, image of the similarity more than 90% is deleted；Next, being carried out to the lesion image of reservation at gray scale
Reason, brightness Y and tri- color components of R, G, B corresponding relation are set up according to the variation relation of RGB and YUV color spaces：Y=
0.3R+0.59G+0.11B, the gray value of image is expressed with this brightness value, final lesion image is obtained and is sent to image
Identification module；
The pathological image for the corresponding focus that is stored with pathology storehouse and healthy image；
The lesion image that picture recognition module exports image processing module is schemed with the pathological image in pathology storehouse and health
As carrying out image recognition, system provides pathological diagnosis result according to image recognition result.
Alternatively, the pathological image in the pathology storehouse and healthy image are the image by pretreatment, preprocessing process
For：Pathological image and healthy image are split into multiple big regions, big region is divided according to this kind of focus easily ill position,
Each big region is divided into many parts of zonules in order again, and there is a sequence pathological image and healthy all zonules of image
Number；
Described image identification module receive image processing module transmission lesion image, according to pathological image in pathology storehouse
Principle being split with healthy image identical lesion image being divided into many parts of zonules, there is a sequence number all zonules；
Lesion image and the zonule of healthy image same sequence number are compared described image identification module.
Alternatively, described image identification module is compared lesion image and the zonule of healthy image same sequence number
It is right, if in comparison result, similarity zonule quantity up to standard is not above standard value, then it is assumed that the lesion image is pathology
Sample；Next, the pathology sample is again compared with different pathological images, when the zonule up to standard of similarity in comparative result
Quantity is higher than standard value, then it is assumed that the lesion image is the pathology.
Alternatively, the split process to lesion image, healthy image, pathological image is specially：
Step (a), mesh generation, the big region of one-level mesh generation, two grades of mesh generation zonules are carried out to image；
Step (b), creates a zonule node set, and a data area, and the data area is used to deposit every
Individual zonule address of node；
Step (c), each one single pixel storage area of zonule address of node correspondence, for depositing the cell
Domain all pixels data；
Step (d), each zonule is traveled through according to sequence number successively, by the pixel of each in zonule Vi (xi,
Yi, zi) the corresponding pixel storage area in the zonule is stored in, after traversal is completed, this image, which is split, to be completed.
Alternatively, the mistake that lesion image is compared described image identification module with healthy image or pathological image
Journey, be specially：
Step (e), the zonule of different images same sequence number is compared, if pixel matching degree reaches standard
Value, then it is assumed that two zonules of the sequence number are similar；
In step (f), big region, the quantity of similar zonule reaches standard value, then it is assumed that the sequence number identical two
Big region is similar；
Step (g), if the similar quantity in big region reaches standard value, then it is assumed that two images match.
The invention also provides a kind of pathological analysis method based on image, comprise the following steps：
Step one, the focus image of lesions position is shot；
Step 2, image procossing is carried out to focus image, first, extracts the random frame figure in every millisecond of image data
Picture, is lesion image by focus video conversion；Then, similarity screening is carried out to lesion image, deletes similarity more than 90%
Image；Next, carrying out gray proces to the lesion image of reservation, set up according to the variation relation of RGB and YUV color spaces
Brightness Y and tri- color components of R, G, B corresponding relation：Y=0.3R+0.59G+0.11B, image is expressed with this brightness value
Gray value, obtains final lesion image；
Step 3, image recognition is carried out by the pathological image in lesion image and pathology storehouse and healthy image, and according to figure
As recognition result provides pathological diagnosis result.
Alternatively, the pathological image in the pathology storehouse and healthy image are the image by pretreatment, preprocessing process
For：Pathological image and healthy image are split into multiple big regions, big region is divided according to this kind of focus easily ill position,
Each big region is divided into many parts of zonules in order again, and there is a sequence pathological image and healthy all zonules of image
Number；
It is many according to the lesion image is divided into pathological image in pathology storehouse and healthy image identical fractionation principle
Part zonule, there is a sequence number all zonules；
Lesion image and the zonule of healthy image same sequence number are compared.
Alternatively, lesion image and the zonule of healthy image same sequence number are compared, if in comparison result,
Similarity zonule quantity up to standard is not above standard value, then it is assumed that the lesion image is pathology sample；Next, the pathology
Sample is again compared with different pathological images, when the zonule quantity up to standard of similarity in comparative result is higher than standard value, then
It is the pathology to think the lesion image.
Alternatively, the split process to lesion image, healthy image, pathological image is specially：
Step (a), mesh generation, the big region of one-level mesh generation, two grades of mesh generation zonules are carried out to image；
Step (b), creates a zonule node set, and a data area, and the data area is used to deposit every
Individual zonule address of node；
Step (c), each one single pixel storage area of zonule address of node correspondence, for depositing the cell
Domain all pixels data；
Step (d), each zonule is traveled through according to sequence number successively, by the pixel of each in zonule Vi (xi,
Yi, zi) the corresponding pixel storage area in the zonule is stored in, after traversal is completed, this image, which is split, to be completed；
Alternatively, process lesion image being compared with healthy image or pathological image, be specially：
Step (e), the zonule of different images same sequence number is compared, if pixel matching degree reaches standard
Value, then it is assumed that two zonules of the sequence number are similar；
In step (f), big region, the quantity of similar zonule reaches standard value, then it is assumed that the sequence number identical two
Big region is similar；
Step (g), if the similar quantity in big region reaches standard value, then it is assumed that two images match.
The beneficial effects of the invention are as follows：
(1) Imaging Technology is combined with medical treatment, by picture recognition module and pathology storehouse to being dispersed in various regions hospital
Image capture device collection lesion image carry out analysis judgement, conditions of patients is tackled in time；
(2) when that can not be judged by digitizing technique, expert consultation is realized by interconnecting expert's computer end, is entered
And effective distribution of medical resource is realized, enable the patient to obtain the remote medical consultation with specialists of expert；
(3) most of state of an illness for being easy to judge are analyzed by pathology storehouse and provides result, while providing corresponding
Therapeutic scheme, the state of an illness that can not be judged for pathology storehouse, by expert by Artificial Diagnosis, saves the time of Medical Technologist, and
The optimum allocation of the high-quality medical resource of realization.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the accompanying drawing used required in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can be with
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is a kind of control block diagram of the pathological analysis system based on image of the present invention；
Fig. 2 carries out the flow chart of split process for the present invention to lesion image, healthy image, pathological image；
For the picture recognition module of the present invention process is compared with healthy image or pathological image in lesion image by Fig. 3
Flow chart.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.It is based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of creative work is not made
Embodiment, belongs to the scope of protection of the invention.
As shown in figure 1, the present invention proposes a kind of digitlization pathological analysis system, including：At image acquisition module, image
Manage module, picture recognition module and pathology storehouse.
The digitlization pathological analysis system of the present invention is distributed in each hospital, and medical under-developed area is realized by the system
Diagnosis to difficult or malignant disease.
Image acquisition module can be camera, and camera is carried by toter and is delivered to the detecting position in human body
Put, the lesion image for shooting lesions position.So, the patient being distributed throughout the country just can be medical in locality, realizes figure
As gathering nearby, patient's medical treatment is greatly facilitated.
The sample position of camera and sampling frequency are set according to the disease locus of focus, the high position of such as frequency of disease development
Sampling frequency is high, and the low position sample frequency of frequency of disease development is low.Or, the sampling of focus is manually gathered completion by doctor, works as hair
Sampled during existing focus, do not find not sample during focus.
Image processing module receives the focus image that camera is shot, and image procossing is carried out to focus image, first, extracts
A random two field picture in every millisecond of image data, is lesion image by focus video conversion；Then, phase is carried out to lesion image
Like degree screening, image of the similarity more than 90% is deleted；Next, gray proces are carried out to the lesion image of reservation, according to RGB
Brightness Y and tri- color components of R, G, B corresponding relation are set up with the variation relation of YUV color spaces：Y=0.3R+0.59G+
0.11B, the gray value of image is expressed with this brightness value, final lesion image is obtained and is sent to picture recognition module.
The pathological image for the corresponding focus that is stored with pathology storehouse and healthy image.
The lesion image that picture recognition module exports image processing module is schemed with the pathological image in pathology storehouse and health
As carrying out image recognition, and the similarity of two images compared is provided, system provides pathology according to image recognition result and examined
Disconnected result, if the similarity of two images is higher than standard value, is diagnosed as the pathology.
The pathological image for the various illnesss that are stored with pathology storehouse and healthy image, because the image of each illness has been carried out
Classification storage, therefore, can be schemed by selection sort with all pathological images of this kind of focus in preliminary latch pathology storehouse and health
Picture.
Pathological image and healthy image in pathology storehouse are the image by pretreatment, and preprocessing process is：By pathology figure
Picture and healthy image split into multiple big regions, and big region is divided according to focus easily ill position, and each big region is pressed again
Order is divided into many parts of zonules, and such pathological image and healthy all zonules of image are all assigned a sequence number.
Picture recognition module receives the lesion image of image processing module transmission, identical with healthy image according to pathological image
Fractionation principle the lesion image that camera is shot is divided into many parts of zonules, there is a sequence number all zonules.
Lesion image and the zonule of healthy image same sequence number are compared picture recognition module, if comparing knot
In fruit, similarity zonule quantity up to standard is not above standard value, then it is assumed that the lesion image is pathology sample, next,
The pathology sample is again compared with different pathological images, when the zonule quantity up to standard of similarity in comparative result is higher than standard
Value, then it is assumed that the pathological image is the pathology.If all pathological images are all mismatched with the lesion image of upload in pathology storehouse,
Lesion image is then distributed to domain expert's computer end, manually providing pathology by the domain expert judges and administering method.
As shown in Fig. 2 the split process to micro image, healthy image, pathological image is specially：
Step (a), mesh generation, one-level grid are carried out to image (micro image or healthy image or pathological image)
Divide big region, two grades of mesh generation zonules；One-level grid is not fixed size, nor fixed shape, due to big
Region is divided according to this kind of focus easily ill position, therefore, and one-level grid draws lesion image according to easily ill position
It is divided into multiple big regions, focus is drawn a circle to approve in one-level grid, then again in one-level grid with two grades of mesh generation zonules；
Step (b), creates a zonule node set, and a data area, and the data area is used to deposit every
Individual zonule address of node；
Step (c), each one single pixel storage area of zonule address of node correspondence, for depositing the cell
Domain all pixels data；
Step (d), each zonule is traveled through according to sequence number successively, by the pixel of each in zonule Vi (xi,
Yi, zi) the corresponding pixel storage area in the zonule is stored in, after traversal is completed, this image, which is split, to be completed.
As shown in figure 3, next, pathological image is compared picture recognition module with healthy image or pathological image
Process, be specially：
Step (e), the zonule of different images same sequence number is compared, if pixel matching degree reaches standard
Value, then it is assumed that two zonules of the sequence number are similar；
In step (f), big region, the quantity of similar zonule reaches standard value, then it is assumed that the sequence number identical two
Big region is similar；
Step (g), if the similar quantity in big region reaches standard value, then it is assumed that two images match.
By above-mentioned comparison process, it can accurately judge the similarity of two images, focus figure is judged according to similarity
Seem health or pathology, if pathology, then which kind of pathology can be matched, and then provide administering method or seek expert people
Work judges.
The pathological analysis system based on image of the present invention, iconography is combined with medical treatment, the intelligence of focus is realized
Diagnosis and effective distribution of Expert Resources, realize the difficult and complicated illness or malignant diseases of outlying district or medical under-developed area
The diagnosis of disease.
The invention also provides one kind digitlization pathological analysis method, its analysis principle is identical with above-mentioned analysis system, this
In repeat no more.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
God is with principle, and any modification, equivalent substitution and improvements made etc. should be included in the scope of the protection.
Claims (10)
1. a kind of pathological analysis system based on image, it is characterised in that including：Image acquisition module, image processing module, figure
As identification module and pathology storehouse；
Image acquisition module is carried by toter and is delivered to the test position in human body, the disease of captured in real-time lesions position
Stove image；
Image processing module receives the focus image that image acquisition module is shot, and carries out image procossing to focus image, first, carries
The random two field picture in every millisecond of image data is taken, is lesion image by focus video conversion；Then, lesion image is carried out
Similarity is screened, and deletes image of the similarity more than 90%；Next, gray proces are carried out to the lesion image of reservation, according to
The variation relation of RGB and YUV color spaces sets up brightness Y and tri- color components of R, G, B corresponding relation：Y=0.3R+
0.59G+0.11B, the gray value of image is expressed with this brightness value, final lesion image is obtained and is sent to image recognition mould
Block；
The pathological image for the corresponding focus that is stored with pathology storehouse and healthy image；
The lesion image that picture recognition module exports image processing module enters with the pathological image in pathology storehouse and healthy image
Row image recognition, system provides pathological diagnosis result according to image recognition result.
2. a kind of pathological analysis system based on image as claimed in claim 1, it is characterised in that
Pathological image and healthy image in the pathology storehouse are the image by pretreatment, and preprocessing process is：By pathology figure
Picture and healthy image split into multiple big regions, and big region is divided according to this kind of focus easily ill position, each big region
It is divided into many parts of zonules in order again, there is a sequence number pathological image and healthy all zonules of image；
Described image identification module receive image processing module transmission lesion image, according to pathological image in pathology storehouse and be good for
Health image identical splits principle and lesion image is divided into many parts of zonules, and there is a sequence number all zonules；
Lesion image and the zonule of healthy image same sequence number are compared described image identification module.
3. a kind of pathological analysis system based on image as claimed in claim 2, it is characterised in that described image identification module
The zonule of lesion image and healthy image same sequence number is compared, if in comparison result, up to standard small of similarity
Region quantity is not above standard value, then it is assumed that the lesion image is pathology sample；Next, the pathology sample again from it is different
Pathological image compares, when the zonule quantity up to standard of similarity in comparative result is higher than standard value, then it is assumed that the lesion image
For the pathology.
4. a kind of pathological analysis system based on image as claimed in claim 2, it is characterised in that it is described to lesion image,
Healthy image, the split process of pathological image are specially：
Step (a), mesh generation, the big region of one-level mesh generation, two grades of mesh generation zonules are carried out to image；
Step (b), creates a zonule node set, and a data area, and the data area is used to deposit each small
The address of Area Node；
Step (c), each one single pixel storage area of zonule address of node correspondence, for depositing the zonule institute
There is pixel data；
Step (d), each zonule is traveled through according to sequence number successively, by the pixel of each in zonule Vi (xi, yi, zi)
The corresponding pixel storage area in the zonule is stored in, after traversal is completed, this image, which is split, to be completed.
5. a kind of pathological analysis system based on image as claimed in claim 3, it is characterised in that
The process that lesion image is compared described image identification module with healthy image or pathological image, be specially：
Step (e), the zonule of different images same sequence number is compared, if pixel matching degree reaches standard value,
Two zonules for thinking the sequence number are similar；
In step (f), big region, the quantity of similar zonule reaches standard value, then it is assumed that sequence number identical Liang Ge great areas
Domain is similar；
Step (g), if the similar quantity in big region reaches standard value, then it is assumed that two images match.
6. a kind of pathological analysis method based on image, it is characterised in that comprise the following steps：
Step one, the focus image of lesions position is shot；
Step 2, image procossing is carried out to focus image, first, extracts the random two field picture in every millisecond of image data, will
Focus video conversion is lesion image；Then, similarity screening is carried out to lesion image, deletes figure of the similarity more than 90%
Picture；Next, carrying out gray proces to the lesion image of reservation, brightness is set up according to the variation relation of RGB and YUV color spaces
Y and tri- color components of R, G, B corresponding relation：Y=0.3R+0.59G+0.11B, the gray scale of image is expressed with this brightness value
Value, obtains final lesion image；
Step 3, carries out image recognition, and know according to image by the pathological image in lesion image and pathology storehouse and healthy image
Other result provides pathological diagnosis result.
7. a kind of pathological analysis method based on image as claimed in claim 6, it is characterised in that
Pathological image and healthy image in the pathology storehouse are the image by pretreatment, and preprocessing process is：By pathology figure
Picture and healthy image split into multiple big regions, and big region is divided according to this kind of focus easily ill position, each big region
It is divided into many parts of zonules in order again, there is a sequence number pathological image and healthy all zonules of image；
It is many parts small according to the lesion image is divided into pathological image in pathology storehouse and healthy image identical fractionation principle
There is a sequence number region, all zonules；
Lesion image and the zonule of healthy image same sequence number are compared.
8. a kind of pathological analysis method based on image as claimed in claim 7, it is characterised in that by lesion image and health
The zonule of image same sequence number is compared, if in comparison result, similarity zonule quantity up to standard is not above
Standard value, then it is assumed that the lesion image is pathology sample；Next, the pathology sample is again compared with different pathological images,
When the zonule quantity up to standard of similarity in comparative result is higher than standard value, then it is assumed that the lesion image is the pathology.
9. a kind of pathological analysis system based on image as claimed in claim 7, it is characterised in that to lesion image, health
Image, the split process of pathological image are specially：
Step (a), mesh generation, the big region of one-level mesh generation, two grades of mesh generation zonules are carried out to image；
Step (b), creates a zonule node set, and a data area, and the data area is used to deposit each small
The address of Area Node；
Step (c), each one single pixel storage area of zonule address of node correspondence, for depositing the zonule institute
There is pixel data；
Step (d), each zonule is traveled through according to sequence number successively, by the pixel of each in zonule Vi (xi, yi, zi)
The corresponding pixel storage area in the zonule is stored in, after traversal is completed, this image, which is split, to be completed.
10. a kind of pathological analysis system based on image as claimed in claim 8, it is characterised in that
The process that lesion image is compared with healthy image or pathological image, be specially：
Step (e), the zonule of different images same sequence number is compared, if pixel matching degree reaches standard value,
Two zonules for thinking the sequence number are similar；
In step (f), big region, the quantity of similar zonule reaches standard value, then it is assumed that sequence number identical Liang Ge great areas
Domain is similar；
Step (g), if the similar quantity in big region reaches standard value, then it is assumed that two images match.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710609478.8A CN107274404A (en) | 2017-07-25 | 2017-07-25 | A kind of pathological analysis system and method based on image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710609478.8A CN107274404A (en) | 2017-07-25 | 2017-07-25 | A kind of pathological analysis system and method based on image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107274404A true CN107274404A (en) | 2017-10-20 |
Family
ID=60079581
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710609478.8A Withdrawn CN107274404A (en) | 2017-07-25 | 2017-07-25 | A kind of pathological analysis system and method based on image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107274404A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108172276A (en) * | 2017-12-01 | 2018-06-15 | 杭州远舟医疗科技有限公司 | A kind of infrared image automatic evaluation method, electronic equipment and storage medium |
| CN108674453A (en) * | 2018-05-23 | 2018-10-19 | 深圳市阡丘越科技有限公司 | A kind of rail traffic monitoring system and control method based on AI technologies |
| CN109953430A (en) * | 2017-12-26 | 2019-07-02 | 金洁 | Computer management system in the production of shoes industry |
| CN110348477A (en) * | 2019-06-04 | 2019-10-18 | 上海联影智能医疗科技有限公司 | Medical image processing method, storage medium and computer equipment |
| CN113990458A (en) * | 2021-12-28 | 2022-01-28 | 深圳市海瑞洋科技有限公司 | Medical electronic endoscope image processing system and method |
| CN114298976A (en) * | 2021-12-08 | 2022-04-08 | 成都道可名康医学检验实验室有限公司 | Medical pathological image diagnosis method |
| CN115797272A (en) * | 2022-11-17 | 2023-03-14 | 普罗朗生物技术（无锡）有限公司 | Lung health detection system based on big data AI |
| CN115965626A (en) * | 2023-03-16 | 2023-04-14 | 邦世科技(南京)有限公司 | Medical image processing method and system based on artificial intelligence algorithm |
| CN116468727A (en) * | 2023-06-19 | 2023-07-21 | 湖南科迈森医疗科技有限公司 | Method and system for assisting in judging high-risk endometrial hyperplasia based on endoscopic image recognition |
-
2017
- 2017-07-25 CN CN201710609478.8A patent/CN107274404A/en not_active Withdrawn
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108172276A (en) * | 2017-12-01 | 2018-06-15 | 杭州远舟医疗科技有限公司 | A kind of infrared image automatic evaluation method, electronic equipment and storage medium |
| CN109953430A (en) * | 2017-12-26 | 2019-07-02 | 金洁 | Computer management system in the production of shoes industry |
| CN108674453A (en) * | 2018-05-23 | 2018-10-19 | 深圳市阡丘越科技有限公司 | A kind of rail traffic monitoring system and control method based on AI technologies |
| CN110348477A (en) * | 2019-06-04 | 2019-10-18 | 上海联影智能医疗科技有限公司 | Medical image processing method, storage medium and computer equipment |
| CN110348477B (en) * | 2019-06-04 | 2021-10-22 | 上海联影智能医疗科技有限公司 | Medical image processing method, storage medium, and computer device |
| CN114298976A (en) * | 2021-12-08 | 2022-04-08 | 成都道可名康医学检验实验室有限公司 | Medical pathological image diagnosis method |
| CN113990458A (en) * | 2021-12-28 | 2022-01-28 | 深圳市海瑞洋科技有限公司 | Medical electronic endoscope image processing system and method |
| CN115797272A (en) * | 2022-11-17 | 2023-03-14 | 普罗朗生物技术（无锡）有限公司 | Lung health detection system based on big data AI |
| CN115965626A (en) * | 2023-03-16 | 2023-04-14 | 邦世科技(南京)有限公司 | Medical image processing method and system based on artificial intelligence algorithm |
| CN116468727A (en) * | 2023-06-19 | 2023-07-21 | 湖南科迈森医疗科技有限公司 | Method and system for assisting in judging high-risk endometrial hyperplasia based on endoscopic image recognition |
| CN116468727B (en) * | 2023-06-19 | 2023-12-12 | 湖南科迈森医疗科技有限公司 | Method and system for assisting in judging high-risk endometrial hyperplasia based on endoscopic image recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107274404A (en) | A kind of pathological analysis system and method based on image | |
| WO2020151536A1 (en) | Brain image segmentation method, apparatus, network device and storage medium | |
| Ozcelik et al. | Reconstruction of perceived images from fmri patterns and semantic brain exploration using instance-conditioned gans | |
| JP4669268B2 (en) | Method and apparatus for detecting and displaying changes in image over time | |
| Pang et al. | Tongue image analysis for appendicitis diagnosis | |
| CN1795272B (en) | System and method for rapidly identifying pathogens, bacteria and abnormal cells | |
| CN107220523A (en) | One kind digitlization pathological analysis system and method | |
| CN110349156A (en) | The recognition methods of lesion characteristics and device, storage medium in the picture of eyeground | |
| CN109615633A (en) | Crohn disease assistant diagnosis system and method under a kind of colonoscopy based on deep learning | |
| CN109948719B (en) | Automatic fundus image quality classification method based on residual dense module network structure | |
| DE112020001683T5 (en) | A PLATFORM FOR THE EVALUATION OF MEDICAL INFORMATION AND METHODS FOR USING THEM | |
| CN106548038A (en) | A kind of Knowledge based engineering medical imaging acquisition analysis system | |
| WO2013084123A2 (en) | Selection of images for optical examination of the cervix | |
| CN106250377A (en) | A kind of method based on the Internet+quick obtaining personalized medicine inspection service | |
| WO2021068857A1 (en) | Artificial intelligence-based cloud diagnosis platform | |
| CN114842957B (en) | Senile dementia auxiliary diagnosis system and method based on emotion recognition | |
| CN107977958A (en) | A kind of image diagnosing method and device | |
| CN113792807B (en) | Skin disease classification model training method, system, medium and electronic equipment | |
| CN116313021B (en) | Medical treatment management system and method for hospital | |
| CN111127391B (en) | Device for dynamically identifying whether nodules are benign or malignant based on thyroid ultrasound video flow | |
| CN111462869A (en) | Multifunctional head image diagnosis and treatment system and method for neonate clinic | |
| US11574726B2 (en) | Image analysis method for multicenter study and system thereof | |
| CN109979588A (en) | Image scanning pathological section system | |
| CN113143201A (en) | Diagnosis system based on tongue coating and tongue quality images | |
| Xue et al. | A unified set of analysis tools for uterine cervix image segmentation |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20171020 |