A classification method for diabetic retinopathy based on multi-scale cascade

A diabetic retina and multi-scale cascading technology, applied in the field of computer vision, can solve problems such as the impact of DR grading results, and achieve the effect of preventing loss and improving performance

Active Publication Date: 2022-04-05
SHENZHEN UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

These small lesions are easily overlooked during the convolution process and will affect the final DR classification results

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 classification method for diabetic retinopathy based on multi-scale cascade
  • A classification method for diabetic retinopathy based on multi-scale cascade
  • A classification method for diabetic retinopathy based on multi-scale cascade

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] refer to figure 1 As shown, the present invention proposes a method for grading diabetic retinopathy based on multi-scale cascade, which specifically includes the following steps:

[0040] S1. Collect labeled diabetic retinopathy fundus images as an original data set; and divide the original data set into a training set and a test set in proportion;

[0041] The original dataset uses the APTOS 2019 Blindness Detection public dataset on kaggle.

[0042]The original data set is divided into a training set and a test set according to a random sampling combination with a ratio of 8:2, so as to reduce the random error of the network.

[0043] S2, construct the Res2Net network model, and set the input batch size parameter of the Res2Net network model;

[0044] S3. According to the input batch size parameter, input the training set into the Res2Net network model for training, and optimize it with an SGD optimizer to obtain a trained Res2Net network model;

[0045] The inpu...

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 method for grading diabetic retinopathy based on multi-scale cascading, including: a Res2Net basic network for extracting multi-scale information of input images, an attention module for extracting feature maps of the first layer, and integrating feelings with more discriminative feature representations The field module and multi-scale cascaded network; the loss of shallow information is reduced by extracting multi-scale information, and the shallow information and high-level information are fused through cascading. Complementarity to enhance the acquisition of information, combined with multi-scale and cascade can effectively improve the grading effect of DR, which has certain clinical and algorithmic significance.

Description

technical field [0001] The invention relates to the fields of computer vision and medical treatment, in particular to a method for grading diabetic retinopathy based on multi-scale cascading. Background technique [0002] Diabetic retinopathy (DR) is a fundus complication caused by microvascular deterioration of diabetes, and eventually causes visual defects or irreversible blindness. The pathological features of diabetic retinopathy mainly include the following categories: microaneurysm, hemorrhage, hard exudate and soft exudate. According to its type and the number of lesions in fundus images, the international clinical diabetic retinopathy grading standard divides diabetic retinopathy into five stages: no obvious diabetic retinopathy, mild non-proliferative diabetic retinopathy, moderate non-proliferative diabetic retinopathy , Severe non-proliferative diabetic retinopathy, proliferative diabetic retinopathy. In the current clinical diagnosis, DR classification mainly r...

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): G06K9/62G06N3/04G06N3/08
CPCG06N3/04G06N3/08G06F18/2414G06F18/253
Inventor 岳广辉李苑汪天富林嘉琪李洁玉周天薇
Owner SHENZHEN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap