Unlock instant, AI-driven research and patent intelligence for your innovation.

Small sample diabetic retinopathy classification system based on model independent element learning

A technology of diabetic retinopathy and classification system, applied in the field of small-sample diabetic retinopathy classification system, can solve problems such as low discrimination, difficulty in classifying lesions, difficulty in collecting images, etc., achieve good classification effect, improve rapid adaptability and accurate classification rate effect

Pending Publication Date: 2021-10-22
SHANDONG NORMAL UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, deep learning methods often require a large amount of training data to train the model
Labeled data are usually limited in real life, and for diabetic retinopathy, it is very difficult to collect a large number of images of various lesion levels
Moreover, due to the interference of light brightness and other factors during image formation, the lesions in the retinal image are sometimes not very distinguishable from the background, and it is even more difficult to classify the degree of 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
  • Small sample diabetic retinopathy classification system based on model independent element learning
  • Small sample diabetic retinopathy classification system based on model independent element learning
  • Small sample diabetic retinopathy classification system based on model independent element learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0051] This embodiment provides a diabetic retinopathy classification system based on model-independent meta-learning, using a small-sample diabetic retinopathy classification model to obtain prior knowledge by learning multiple similar tasks, and using the prior knowledge in the meta-test stage Achieving fast and accurate small-sample diabetic retinopathy classification.

[0052] combine figure 1 and figure 2 , the system specifically includes:

[0053] The data acquisition module is configured to: acquire diabetic retinopathy images to be classified;

[0054] The image classification module is configured to: classify the diabetic retinopathy image to be classified based on a small-sample diabetic retinopathy classification model;

[0055] Wherein, the small-sample diabetic retinopathy classification model is obtained based on the model training module, and the model training module specifically includes:

[0056] The data set acquisition module is configured to: acquire d...

Embodiment 2

[0083] The purpose of this embodiment is to provide a small-sample diabetic retinopathy classification model training method based on model-independent meta-learning, such as image 3 shown, including the following steps:

[0084] Obtain diabetic retinopathy data containing multiple categories, construct a meta-training set based on some of the categories of data, and construct a meta-test set based on other categories of data;

[0085] The meta-training set was used to meta-train the pre-built convolutional neural network model; the meta-test set was used to test the model and adjust parameters; finally, a small-sample diabetic retinopathy classification model was obtained.

Embodiment 3

[0087] The purpose of this embodiment is to provide a small-sample diabetic retinopathy classification model training system based on model-independent meta-learning, including:

[0088] The data set acquisition module is configured to: acquire diabetic retinopathy data including multiple categories, construct a meta-training set based on some of the categories of data, and construct a meta-test set based on other categories of data;

[0089] The meta-learning module is configured to: use the meta-training set to perform meta-training on the pre-built convolutional neural network model; use the meta-testing set to test the model and adjust parameters; and finally obtain a small-sample diabetic retinopathy classification model.

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 small sample diabetic retinopathy classification system based on model-independent meta-learning, and the system comprises: a data acquisition module which is configured to obtain a to-be-classified diabetic retinopathy image; an image classification module which is configured to classify the to-be-classified diabetic retinopathy images based on a small sample diabetic retinopathy classification model, wherein the classification model is obtained through training by adopting a model-independent meta-learning method, and a loss function in the model adopts a difficult task perception loss function. According to the system, the dependence of the model on the sample number is effectively reduced, and the classification accuracy of the small sample diabetic retinopathy is improved.

Description

technical field [0001] The invention belongs to the field of machine learning, and in particular relates to a small-sample diabetic retinopathy classification system based on difficult task perception model-independent element learning. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art. [0003] Diabetic retinopathy is one of the leading causes of blindness. The classification of color fundus images is of great significance to the prevention and treatment of diabetic retinopathy. Effective computer-aided classification technology can greatly save ophthalmologists' diagnosis time and improve the efficiency and accuracy of diabetic retinopathy classification. [0004] In recent years, research on the classification of diabetic retinopathy based on deep learning has made great progress. However, deep learning methods often require a large amount of t...

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): G06T7/00G06K9/62G06N3/04G06N3/08
CPCG06T7/0012G06N3/08G06T2207/20081G06T2207/20084G06T2207/30041G06T2207/30096G06N3/045G06F18/24G06F18/214
Inventor 李登旺董雪媛黄浦刘学尧姜泽坤宋卫清高祝敏陈美荣薛洁赵睿
Owner SHANDONG NORMAL UNIV