Retinal Vessel Segmentation Method Fused with w-net and Conditional Generative Adversarial Network

A technique for generating retinal blood vessels and conditions, applied in biological neural network models, image analysis, image enhancement, etc., can solve the problems of insufficient microvascular segmentation, over-segmentation, low sensitivity, etc. excellent effect

Active Publication Date: 2022-04-19
JIANGXI UNIV OF SCI & TECH
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to propose a retinal vessel segmentation method that integrates W-net and conditional generative adversarial network in view of the problems of low sensitivity, insufficient or over-segmented microvessel segmentation in existing retinal vessel segmentation algorithms

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
  • Retinal Vessel Segmentation Method Fused with w-net and Conditional Generative Adversarial Network
  • Retinal Vessel Segmentation Method Fused with w-net and Conditional Generative Adversarial Network
  • Retinal Vessel Segmentation Method Fused with w-net and Conditional Generative Adversarial Network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0041] The present invention expands U-net to W-net, and uses depth-separable convolution and residual modules in W-net to avoid gradient disappearance due to too deep network, introduces SE module, and distributes weights to each channel , so as to ensure that important features are fully learned, avoid learning useless features, and integrate W-net with conditional generation confrontation network, which can make full use of the strong learning ability of W-net for microvascular features and the strong discrimination ability of CGAN for microvascular features. Extract as many microvessels as possible while ensuring the complete extraction of main vessels. The invention has the advantages of high retinal blood vessel segmentation accuracy and low model complexity, can be used as a computer-aided diagnosis system, improves the doctor's diagnosis efficiency, reduces the misdiagnosis rate, and saves precious time of patients.

[0042] Experiment description: The example data com...

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 relates to the application of deep learning algorithms in the field of medical image analysis, in particular to a retinal vessel segmentation algorithm that integrates W-net and conditional generation confrontation network. The present invention better solves the problems of low segmentation sensitivity and insufficient segmentation of microvessels, and achieves great progress in network parameter utilization, information flow and feature resolution, and contributes to complete segmentation of main vessels and fine segmentation of microvessels , and the intersection of blood vessels is not easy to break, and the lesion and optic disc are not easy to be mistakenly divided into blood vessels. The present invention integrates multiple network models with relatively low complexity, and has excellent overall segmentation performance on the DRIVE data set, with sensitivity and accuracy rates of 87.18% and 96.95% respectively, and a ROC curve value of 98.42%, which can be used in medical treatment Computer-aided diagnosis in the field, enabling rapid and automated retinal vessel segmentation.

Description

technical field [0001] The invention relates to the application of deep learning algorithms in the field of medical image analysis, in particular to a retinal vessel segmentation algorithm that integrates W-net and conditional generation confrontation network. Background technique [0002] Diabetic retinopathy, cardiovascular disease, hypertension, arteriosclerosis and other diseases have different effects on retinal blood vessels, which can be diagnosed by analyzing the characteristics of blood vessels in retinal fundus images such as length, width, angle, curvature and branch form. In order to obtain a more accurate pathological diagnosis, retinal vessels must be accurately segmented from fundus images, and manual segmentation of retinal vessels is a cumbersome, complex and highly professional task, and the segmentation standards are highly subjective. In order to improve the doctor's diagnostic efficiency and reduce the misdiagnosis rate, a computer-aided diagnosis system...

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): G06T7/11G06N3/04
CPCG06T7/11G06T2207/20084G06T2207/30101G06N3/048
Inventor 梁礼明蓝智敏吴健盛校棋杨国亮冯新刚
Owner JIANGXI UNIV OF SCI & TECH
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