CN110459299B - Retina fundus color photograph image screening method - Google Patents
Retina fundus color photograph image screening method Download PDFInfo
- Publication number
- CN110459299B CN110459299B CN201910619610.2A CN201910619610A CN110459299B CN 110459299 B CN110459299 B CN 110459299B CN 201910619610 A CN201910619610 A CN 201910619610A CN 110459299 B CN110459299 B CN 110459299B
- Authority
- CN
- China
- Prior art keywords
- image
- images
- similarity
- attribute information
- color photograph
- 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
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
- A61B3/12—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for looking at the eye fundus, e.g. ophthalmoscopes
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
- A61B3/14—Arrangements specially adapted for eye photography
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
-
- 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
- G06T2207/30041—Eye; Retina; Ophthalmic
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention belongs to the technical field of medical image processing, and relates to a method for processing a large-scale retina fundus color photograph image by using an automatic script to screen the retina fundus color photograph image from an appearance image containing a large number of eyes. The invention screens out high-quality fundus images from the large-scale retina fundus color photograph image data by utilizing an image data screening method, allows a user to modify an image similarity calculation area and whether to select a segmentation image, and finally obtains the high-quality retina fundus color photograph image data which can be used for clinical diagnosis and artificial intelligence algorithm development.
Description
Technical Field
The invention relates to the technical field of medical image processing, in particular to a retina fundus color photograph image screening method.
Background
In recent years, with the improvement of the living standard of people, the prevalence of fundus diseases is continuously increased by high-intensity life, unhealthy life style and gradually increasing working pressure, and the eyeground diseases gradually develop to younger. In addition, teenagers have grown in the years due to poor learning habits. The sudden and severe nature of these ophthalmic diseases has severely endangered human lives. Therefore, it is very important to diagnose the related ophthalmic diseases based on the retinal fundus color photograph-early findings and correct diagnosis.
Currently, the relevant diagnosis of fundus lesions is obtained by an ophthalmologist reading each fundus image. Therefore, a high-quality fundus image plays a very important role in clinical diagnosis, for example, further improving the level of ophthalmic diagnosis and treatment, enabling patients to enjoy higher-quality medical services, and the like. Meanwhile, with the rapid development of computer technology, analysis of fundus images by using artificial intelligence technology becomes a research hotspot in the medical field, and as a great amount of data is needed as a basis for an artificial intelligence related algorithm model, how to rapidly obtain high-quality data becomes an important reason for limiting the improvement of the algorithm model efficiency.
Because the quality of various fundus image acquisition devices is uneven at present, image shooting personnel lack training, the obtained images cannot be ensured in quality although the quantity is large, and further, the images cannot be used for developing an artificial intelligence algorithm, so that the development of medical artificial intelligence research is slow. Therefore, how to screen high-quality image data available for diagnosis by doctors and development of artificial intelligence algorithm models from large-scale retinal fundus color photograph image data is also a big problem.
Disclosure of Invention
Aiming at the problem that the prior art can not screen high-quality image data from large-scale retina fundus color photograph image data, the invention provides a retina fundus color photograph image screening method,
a retinal fundus color photograph image screening method, the retinal fundus color photograph image screening method comprising the steps of:
s1, preprocessing retina fundus color photograph image data acquired by different equipment in batches;
s2, selecting an image from the preprocessed image data as a standard template image, and performing image similarity calculation on the image to be screened and the standard template image;
s3, setting a similarity threshold according to the image similarity calculation result, screening out images meeting the requirements according to the similarity threshold, and obtaining attribute information of the images;
s4, matching attribute information of the images meeting the similarity threshold with attribute information of all image data to finish screening of retina fundus color photo images.
Further, in S1, batch preprocessing is performed on the retinal fundus color photograph image data collected by different devices, including but not limited to: image gray level conversion, z-score standardization, contrast-limited self-adaptive histogram equalization, gamma correction, data normalization and the like.
Further, the specific steps of S2 are as follows:
s21, allowing a user to select one image from all the preprocessed images to serve as a standard template image, wherein the difference between image data exists in a specific pixel range, so that the pixel range of the contrast similarity is required to be selected, and a similarity calculation region is set; meanwhile, when the user performs similarity comparison, selecting whether to cut the image in the set range and generating cutting information, namely dividing the calculation area into image blocks with different sizes;
s22, according to the setting of the image similarity calculation area by a user, cutting information for cutting the calculation area is simultaneously applied to the images to be screened, the calculation area is cut into image blocks with different sizes, and the image similarity of the image blocks of the images to be screened and the image blocks of the standard template image is calculated sequentially, so that the similarity index of the images to be screened is obtained.
Further, the step S3 includes setting a threshold according to the image similarity calculation result, screening out images meeting the requirements, and obtaining attribute information of the images, and specifically includes the following steps: and setting a similarity threshold based on a calculation result of the similarity of the images, wherein the images with the similarity index larger than the similarity threshold are images to be screened, and screening the images conforming to the similarity threshold to obtain attribute information of the images conforming to the similarity threshold. The attribute information of the image comprises the name and format information of the image.
Further, the specific steps of S4 are as follows: and matching the attribute information of the image meeting the similarity threshold with the attribute information of all the image data, extracting the image which is the same as the attribute information of the image meeting the similarity threshold, and finishing the screening of the retina fundus color photograph image.
Compared with the prior art, the technical scheme of the invention has the beneficial effects that:
the invention screens out high-quality fundus images from the large-scale retina fundus color photograph image data by utilizing an image data screening method, allows a user to modify an image similarity calculation area and whether to select a cutting image, and finally obtains the high-quality retina fundus color photograph image data which can be used for clinical diagnosis and artificial intelligence algorithm development.
Drawings
FIG. 1 is a flowchart of a method for screening a color photographic image of a fundus oculi according to a preferred embodiment of the present invention;
FIG. 2 is a color photographic image of the retinal fundus acquired;
FIG. 3 is an eye-appearance image;
fig. 4 is a retinal fundus image after pretreatment;
fig. 5 shows a user sketching similarity calculation area, and a is an area size that can be set by the user.
Detailed Description
The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only some embodiments of the present invention, which are only for illustration and not to be construed as limitations of the present patent. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The technical scheme of the invention is further described below with reference to the accompanying drawings and examples.
Example 1
Referring to fig. 1, a flowchart of a method for screening a color photograph of a fundus oculi according to a preferred embodiment of the invention is shown. As shown in fig. 1, the retinal fundus color photograph image screening method includes the steps of:
s1, preprocessing retina fundus color photograph image data acquired by different equipment in batches;
s2, selecting an image from the preprocessed image data as a standard template image, and performing image similarity calculation on the image to be screened and the standard template image;
s3, setting a similarity threshold according to the image similarity calculation result, screening out images meeting the requirements according to the similarity threshold, and obtaining attribute information of the images;
s4, matching the attribute information of the images meeting the similarity threshold value with the attribute information of all the image data, extracting the images identical to the attribute information of the images meeting the similarity threshold value, and finishing screening the retina fundus color photograph images.
Example 2
The method of screening a retinal fundus color photograph image provided in this embodiment is the same as that of embodiment 1, and only the respective steps will be described in further detail.
The scheme adopted by the invention is as follows:
s1, firstly, collecting retina fundus color photograph image data shot by different equipment, and preprocessing the collected data, wherein the preprocessing comprises the steps of but not limited to image gray level conversion, z-score standardization, contrast-limited self-adaptive histogram equalization, gamma correction, data normalization and the like. Fig. 2 is a captured retinal fundus color photograph image, fig. 3 is an eye appearance image, and fig. 4 is a retinal fundus image after preprocessing.
S21, selecting one piece of preprocessed image data as a standard template image, setting a similarity calculation area and selecting whether to divide the image or not, wherein the specific steps are as follows: from all the preprocessed images, allowing a user to select one image from the preprocessed images as a standard template image, and setting a similarity calculation area because differences between image data exist in a specific pixel range, selecting a pixel range for comparing similarity; meanwhile, when the user can select to compare the similarity, whether the image in the set range needs to be cut or not is judged, cutting information is generated, the calculation area is divided into image blocks with different sizes, the similarity calculation area is sketched for the user in fig. 5, and the area size which can be set by the user is A.
S22, according to the setting of the image similarity calculation area by a user, cutting information for cutting the calculation area is simultaneously applied to the images to be screened, the calculation area is cut into image blocks with different sizes, and the image similarity of the image blocks of the images to be screened and the image blocks of the standard template image is calculated sequentially, so that the similarity index of the images to be screened is obtained.
S3, setting a threshold according to the image similarity calculation result, screening out images meeting the requirements and obtaining attribute information of the images, wherein the method specifically comprises the following steps: and setting a similarity threshold based on a calculation result of the similarity of the images, wherein the images with the similarity index larger than the similarity threshold are images to be screened, and screening the images conforming to the similarity threshold to obtain attribute information of the images conforming to the similarity threshold. The attribute information of the image comprises the name and format information of the image.
S4, matching the image names in all the image data by using the attribute information of the images meeting the similarity threshold, extracting the images with the same attribute information as the images meeting the similarity threshold, and finishing the screening of the retina fundus color photograph images.
It is to be understood that the above examples of the present invention are provided by way of illustration only and not by way of limitation of the embodiments of the present invention. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary here nor is it exhaustive of all embodiments. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the invention are desired to be protected by the following claims.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Claims (1)
1. A method for screening a retina fundus color photograph image, which is characterized by comprising the following steps:
s1, preprocessing retina fundus color photograph image data acquired by different equipment in batches;
s2, selecting an image from the preprocessed image data as a standard template image, and performing image similarity calculation on the image to be screened and the standard template image;
s3, setting a similarity threshold according to the image similarity calculation result, screening out images meeting the requirements according to the similarity threshold, and obtaining attribute information of the images;
s4, matching attribute information of the images meeting the similarity threshold with attribute information of all image data to finish screening of retina fundus color photo images;
the specific steps of the S1 are as follows: the retina fundus color photograph image data sequentially performs image gray level conversion, z-score standardization, contrast-limited self-adaptive histogram equalization, gamma correction and data normalization;
the specific steps of the S2 are as follows:
s21, allowing a user to select one image from all the preprocessed images as a standard template image, and setting an image similarity calculation area; meanwhile, when the user selects to compare the image similarity, judging whether the image in the set range needs to be cut and generating corresponding cutting information, namely dividing the calculation area into image blocks with different sizes;
s22, according to the setting of a user on an image similarity calculation area, cutting information for cutting the calculation area is simultaneously applied to an image to be screened, the calculation area is cut into image blocks with different sizes, and the image similarity of the image blocks of the image to be screened and the image blocks of a standard template image is sequentially calculated to obtain a similarity index of the image to be screened;
the specific steps of the S3 are as follows: setting a similarity threshold based on a calculation result of the similarity of the images, wherein the images with the similarity index larger than the similarity threshold are images to be screened, and screening the images conforming to the similarity threshold to obtain attribute information of the images conforming to the similarity threshold;
the attribute information of the image includes: name and format information of the image;
the specific steps of the S4 are as follows: and matching the attribute information of the image meeting the similarity threshold with the attribute information of all the image data, extracting the image which is the same as the attribute information of the image meeting the similarity threshold, and finishing the screening of the retina fundus color photograph image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910619610.2A CN110459299B (en) | 2019-07-10 | 2019-07-10 | Retina fundus color photograph image screening method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910619610.2A CN110459299B (en) | 2019-07-10 | 2019-07-10 | Retina fundus color photograph image screening method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110459299A CN110459299A (en) | 2019-11-15 |
| CN110459299B true CN110459299B (en) | 2023-06-27 |
Family
ID=68482590
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910619610.2A Active CN110459299B (en) | 2019-07-10 | 2019-07-10 | Retina fundus color photograph image screening method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110459299B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111986785B (en) * | 2020-08-26 | 2023-09-12 | 北京至真互联网技术有限公司 | Medical image labeling method, device, equipment and storage medium |
| CN116363132B (en) * | 2023-06-01 | 2023-08-22 | 中南大学湘雅医院 | Ophthalmic image processing method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103870838A (en) * | 2014-03-05 | 2014-06-18 | 南京航空航天大学 | Eye fundus image characteristics extraction method for diabetic retinopathy |
| CN106530295A (en) * | 2016-11-07 | 2017-03-22 | 首都医科大学 | Fundus image classification method and device of retinopathy |
| CN107423571A (en) * | 2017-05-04 | 2017-12-01 | 深圳硅基仿生科技有限公司 | Diabetic retinopathy identifying system based on eye fundus image |
| CN108416371A (en) * | 2018-02-11 | 2018-08-17 | 艾视医疗科技成都有限公司 | A kind of diabetic retinopathy automatic testing method |
| CN108629769A (en) * | 2018-05-02 | 2018-10-09 | 山东师范大学 | Eye fundus image optic disk localization method and system based on best fraternal similarity |
| CN109166117A (en) * | 2018-08-31 | 2019-01-08 | 福州依影健康科技有限公司 | A kind of eye fundus image automatically analyzes comparison method and a kind of storage equipment |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106780439B (en) * | 2016-11-29 | 2019-05-28 | 瑞达昇医疗科技(大连)有限公司 | A method of screening eye fundus image |
| CN107832452A (en) * | 2017-11-23 | 2018-03-23 | 苏州亿科赛卓电子科技有限公司 | A kind of photo management method and device |
| CN108229574B (en) * | 2018-01-18 | 2021-08-03 | 维沃移动通信有限公司 | Picture screening method and device and mobile terminal |
| CN109993079A (en) * | 2019-03-19 | 2019-07-09 | 嘉兴智设信息科技有限公司 | A kind of indoor design picture recognition and screening technique and its device |
-
2019
- 2019-07-10 CN CN201910619610.2A patent/CN110459299B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103870838A (en) * | 2014-03-05 | 2014-06-18 | 南京航空航天大学 | Eye fundus image characteristics extraction method for diabetic retinopathy |
| CN106530295A (en) * | 2016-11-07 | 2017-03-22 | 首都医科大学 | Fundus image classification method and device of retinopathy |
| CN107423571A (en) * | 2017-05-04 | 2017-12-01 | 深圳硅基仿生科技有限公司 | Diabetic retinopathy identifying system based on eye fundus image |
| CN108416371A (en) * | 2018-02-11 | 2018-08-17 | 艾视医疗科技成都有限公司 | A kind of diabetic retinopathy automatic testing method |
| CN108629769A (en) * | 2018-05-02 | 2018-10-09 | 山东师范大学 | Eye fundus image optic disk localization method and system based on best fraternal similarity |
| CN109166117A (en) * | 2018-08-31 | 2019-01-08 | 福州依影健康科技有限公司 | A kind of eye fundus image automatically analyzes comparison method and a kind of storage equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110459299A (en) | 2019-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11666210B2 (en) | System for recognizing diabetic retinopathy | |
| US20220076420A1 (en) | Retinopathy recognition system | |
| US7474775B2 (en) | Automatic detection of red lesions in digital color fundus photographs | |
| CN108198184B (en) | Method and system for vessel segmentation in contrast images | |
| CN107563434B (en) | Brain MRI image classification method and device based on three-dimensional convolutional neural network | |
| CN108765392B (en) | Digestive tract endoscope lesion detection and identification method based on sliding window | |
| CN111489328B (en) | Fundus image quality evaluation method based on blood vessel segmentation and background separation | |
| CN113768461B (en) | Fundus image analysis method, fundus image analysis system and electronic equipment | |
| CN110459299B (en) | Retina fundus color photograph image screening method | |
| CN106780439B (en) | A method of screening eye fundus image | |
| CN113889267A (en) | Method for constructing diabetes diagnosis model based on eye image recognition and electronic equipment | |
| CN111839455A (en) | Eye sign identification method and equipment for thyroid-associated ophthalmopathy | |
| Nayak et al. | Automatic identification of diabetic maculopathy stages using fundus images | |
| CN112233087A (en) | Artificial intelligence-based ophthalmic ultrasonic disease diagnosis method and system | |
| Li et al. | Vessel recognition of retinal fundus images based on fully convolutional network | |
| Lim et al. | Generative data augmentation for diabetic retinopathy classification | |
| CN115409764A (en) | Multi-mode fundus blood vessel segmentation method and device based on domain self-adaptation | |
| Wisaeng et al. | Automatic detection of exudates in retinal images based on threshold moving average models | |
| CN111242850B (en) | Wide-area fundus optical coherence blood flow imaging resolution improving method | |
| CN113408593A (en) | Diabetic retinopathy image classification method based on improved ResNeSt convolutional neural network model | |
| CN116452571A (en) | Image recognition method based on deep neural network | |
| CN111292285A (en) | Automatic screening method for diabetes mellitus based on naive Bayes and support vector machine | |
| Raman et al. | The effects of spatial resolution on an automated diabetic retinopathy screening system's performance in detecting microaneurysms for diabetic retinopathy | |
| Palanisamy et al. | An adaptive enhancement method for low contrast color retinal images based on strucural similarity | |
| Shon et al. | Development of a β-variational autoencoder for disentangled latent space representation of anterior segment optical coherence tomography images |
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 |