Diabetic retinopathy classification system based on uncertainty

A technology of diabetic retina and classification system, applied in the field of diabetic retinopathy classification system, can solve the problem of inability to display the credibility of model classification results, and achieve the effect of safe and reliable clinical use and guaranteed safety performance.

Pending Publication Date: 2020-12-01
中科泰明(南京)科技有限公司
View PDF0 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] Aiming at the deficiencies of the prior art, the present invention provides an uncertainty-based diabetic retinopathy classification system, which solves the problem that the prior art cannot show the credibility of the model 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
  • Diabetic retinopathy classification system based on uncertainty
  • Diabetic retinopathy classification system based on uncertainty
  • Diabetic retinopathy classification system based on uncertainty

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0064] Such as figure 1 As shown, the present invention provides a diabetic retinopathy classification system based on uncertainty, the system includes: a storage module for storing images, a preprocessing module, an exudation segmentation module, a microvascular tumor segmentation module, and a classification module;

[0065] The preprocessing module is used for preprocessing the fundus image; it is also used for normalizing the fused multi-channel image of the exudation segmented image, the microvascular tumor segmented image and the corresponding preprocessed fundus image;

[0066] The exudation segmentation module is used to segment the fundus image into exudation segmentation images through the trained exudation segmentation network model;

[0067] The microvascular tumor segmentation module is used to segment the fundus image into microvascular tumor segmentation images through the trained microvascular tumor segmentation network model;

[0068] The classification modul...

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 provides a diabetic retinopathy classification system based on uncertainty, and relates to the technical field of neural networks. The method includes obtaining an exudation segmentationimage and a micro-hemangioma segmentation image by using the two trained segmentation network models; fusing the exudation segmentation image, the micro-hemangioma segmentation image and the corresponding preprocessed fundus image into a multi-channel image, extracting features by using a trained Bayesian deep learning classification network model, and giving out accidental uncertainty and modeluncertainty by a classification module while finally outputting a model classification result. Therefore, the safety performance of the model is guaranteed, and when a diagnosis image cannot give a result very definitely, whether artificial experts are needed for re-diagnosis or not can be determined through two uncertainty degrees, so that the model is safer and more reliable in clinical use.

Description

technical field [0001] The invention relates to the technical field of neural networks, in particular to an uncertainty-based diabetic retinopathy classification system. Background technique [0002] In recent years, with the increasingly mature development of artificial intelligence, algorithms represented by deep learning have shown great advantages in many medical image applications and have been widely used. For example, using deep learning to detect diabetic retinopathy. [0003] Existing detection methods for diabetic retinopathy usually use convolutional neural network models to detect fundus images as input, and finally obtain the classification results of the model. [0004] However, in medical image processing, methods based on deep learning are very dependent on the quantity and quality of the training dataset of the model, and the parameters of the model are determined after the training. When the model is trained and applied to diabetic retinopathy, the model ...

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 Applications(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08G06T5/50G06T7/10
CPCG06T5/50G06N3/08G06T7/10G06T2207/20221G06T2207/30101G06T2207/30096G06T2207/30041G06N3/045G06F18/24155G06F18/214
Inventor 刘磊
Owner 中科泰明(南京)科技有限公司
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