A multi-lesion image segmentation method for retinal macular edema

A macular edema and image segmentation technology, applied in the field of image processing, can solve the problem of simultaneous segmentation of macular edema images with multiple lesions and multiple lesions, so as to solve the problem of huge data imbalance, improve the segmentation performance, and alleviate the huge data inconsistency. balanced effect

Active Publication Date: 2021-10-22
SUZHOU UNIV
View PDF0 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 overcome the deficiencies in the prior art, provide a method for segmenting images of retinal macular edema and multiple lesions, and solve the technical problem in the prior art that multiple lesions cannot be simultaneously segmented for images of retinal macular edema and multiple lesions

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 multi-lesion image segmentation method for retinal macular edema
  • A multi-lesion image segmentation method for retinal macular edema
  • A multi-lesion image segmentation method for retinal macular edema

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0057] The embodiment of the present invention provides a method for image segmentation of retinal macular edema with multiple lesions. The pre-built and trained codec attention network model is used to realize the joint segmentation of retinal macular edema and multiple lesions, and the results of segmented images are obtained, which can predict macular edema. Multiple lesions in multi-lesion images are segmented simultaneously, laying the foundation for subsequent quantitative analysis of lesions. The present invention will be further described below from image preprocessing, network model design and construction, and experimental comparison results in conjunction with the accompanying drawings. The following examples are only used to illustrate the technical solution of the present invention more clearly, but not to limit the protection scope of the present invention.

[0058] 1. Image preprocessing:

[0059] To prevent GPU memory overflow, the 3D retinal OCT volume data i...

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 image segmentation of retinal macular edema and multiple lesions. The method includes collecting three-dimensional retinal OCT volume data and converting it into a two-dimensional retinal OCT B-scanning image; inputting the two-dimensional retinal OCT B-scanning image into a trained image The encoder-decoder attention network model performs joint segmentation of macular edema and multiple lesions, and obtains the segmentation image results. The invention adopts the codec attention network model to perform joint segmentation of multiple lesions, can obtain more abundant global features, and realizes simultaneous segmentation of retinal edema, subretinal fluid and pigment epithelium detachment multiple lesions in an image of retinal macular edema and multiple lesions. It lays the foundation for the quantitative analysis of subsequent lesions.

Description

technical field [0001] The invention relates to a method for segmenting images of retinal macular edema with multiple lesions, and belongs to the technical field of image processing. Background technique [0002] According to statistics, about 34% of diabetic patients will develop diabetic retinopathy (referred to as sugar net, Diabet Retinopathy, DR). Diabetic reticulum is a common blinding disease. Severe diabetic reticulum disease is often accompanied by retinal edema (Retinal Edema, RE), sub-retinal fluid (Sub-Retinal Fluid, SRF) and pigment epithelial detachment (Pigment epithelial detachment) in the macular area. Epithelial Detachment, PED) and other symptoms. The quantitative segmentation of retinal edema, subretinal fluid and pigment epithelial layer detachment can provide important reference information for clinical analysis and research of diabetic reticulum. However, simultaneous segmentation of multiple lesion regions in Optical Coherence Tomography (OCT) image...

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): G06T7/10G06T7/11G06T3/00
CPCG06T3/0037G06T2207/10012G06T2207/10101G06T2207/20081G06T2207/30041G06T2207/30096G06T7/10G06T7/11
Inventor 朱伟芳冯爽朗陈新建赵鹤鸣
Owner SUZHOU 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