Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A glaucoma medical image classification method based on capsule theory

A medical image and classification method technology, applied in the field of glaucoma medical image analysis, can solve the problem of low accuracy of glaucoma classification, achieve good training stability, fast convergence, and excellent accuracy

Active Publication Date: 2022-04-08
GUANGDONG POLYTECHNIC NORMAL UNIV
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, there is no relevant research on the use of capsule theory to assist glaucoma screening
In addition, directly using the existing deep learning method to process glaucoma medical images, ignoring the unique knowledge in the field of glaucoma, such as the optic disc and cup segmentation map information of the fundus map, will easily lead to low glaucoma classification accuracy

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A glaucoma medical image classification method based on capsule theory
  • A glaucoma medical image classification method based on capsule theory
  • A glaucoma medical image classification method based on capsule theory

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0042] A glaucoma medical image classification method based on capsule theory, comprising the following steps:

[0043] S1: Preprocessing the image;

[0044] S2: Construct a capsule-based convolutional neural network and input the preprocessed image, and output the glaucoma recognition result;

[0045] S3: Build an image reconstruction fully connected network, and use the category capsule to restore the original image.

[0046] In this embodiment, step S1 includes the following steps:

[0047] S1.1: Use the existing glaucoma medical imaging optic disc detection model to cut out the region of interest to reduce interference information;

[0048] S1.2: Fusion of optic disc and optic cup semantic segmentation map information;

[0049]S1.3: Adaptive histogram enhancement with limited contrast is used to improve the overall or local contrast of the image.

[0050] In this embodiment, the capsule-based convolutional neural network described in step S2 includes the first layer of...

Embodiment 2

[0067] Given a fundus map and its optic disc and cup segmentation map annotation information, the image preprocessing technology is used to process it as the input of the trained semantic segmentation network, and the classification probability is output after calculation to determine whether the original image is glaucoma. Process such as Figure 5 shown.

[0068] S1: Read the annotation information of the fundus map and its optic disc and cup segmentation map;

[0069] S2: Fusion of the fundus image and its segmentation image, the original 3-channel RGB fundus image and the single-channel segmentation image are superimposed into a new 4-channel image;

[0070] S3: Carry out the cropping operation of the optic disc area on the new 4-channel image to obtain a region of interest with an image size of 4×28×28;

[0071] S4: Use CLAHE to process the region of interest to enhance image or local contrast;

[0072] S5: use the preprocessed image as the input of the trained capsule...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a glaucoma medical image classification method based on the capsule theory, including image preprocessing operations, constructing a capsule-based convolutional neural network and constructing an image reconstruction fully connected network; the image preprocessing operation first utilizes the existing glaucoma medical The image disc detection model cuts out the region of interest to reduce the interference information, then fuses the optic disc and cup semantic segmentation map information, and finally adopts the adaptive histogram enhancement with limited contrast to improve the overall or local contrast of the image. The capsule-based convolutional neural network can not only automatically learn the features of glaucoma medical images, but also discover the position and direction information between features, so as to identify glaucoma more accurately. Image Reconstruction Fully Connected Network uses category capsules to restore original glaucoma medical images, with the aim of improving the generalization ability of capsule-based convolutional neural networks.

Description

technical field [0001] The present invention relates to the field of glaucoma medical image analysis, and more specifically, relates to a glaucoma medical image classification method based on capsule theory. Background technique [0002] Glaucoma is a group of diseases characterized by optic atrophy, visual field defect and vision loss. If left untreated, it can lead to blindness. Glaucoma has become one of the three major blind diseases, with an incidence rate of 1% in the general population and 2% after the age of 45. Early screening of glaucoma is of great significance, which is conducive to timely treatment of patients, so as to protect vision and even recover. [0003] Due to the large number of potential patients with glaucoma, and the limited medical resources and clinical experience are difficult to meet the existing needs, it is very difficult for ophthalmologists to carry out early screening work. The use of computer-assisted early screening of glaucoma can effe...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06V10/764G06V10/82G06V10/26G06K9/62
CPCG06V10/267G06V2201/03G06F18/24
Inventor 刘少鹏贾西平关立南高维奇洪佳明李耿鑫张倩林智勇崔怀林
Owner GUANGDONG POLYTECHNIC NORMAL UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com