Diabetic retinopathy image classification method based on improved ResNeSt convolutional neural network model

A convolutional neural network and retinopathy technology, applied in the field of medical image processing, can solve the problems of uneven diagnosis level, time-consuming and labor-intensive efficiency, misdiagnosis, etc., achieve good classification results, improve accuracy, and reduce parameters.

Pending Publication Date: 2021-09-17
GUILIN UNIV OF ELECTRONIC TECH +1
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The traditional detection of diabetic retinopathy mainly relies on professional ophthalmologists for manual identification, but this process is time-consuming, labor-intensive

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 image classification method based on improved ResNeSt convolutional neural network model
  • Diabetic retinopathy image classification method based on improved ResNeSt convolutional neural network model
  • Diabetic retinopathy image classification method based on improved ResNeSt convolutional neural network model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] Below in conjunction with accompanying drawing and embodiment the present invention is described in further detail:

[0047] A kind of diabetic retinopathy image classification method based on improved ResNeSt convolutional neural network model, specifically comprises the following steps:

[0048] Step 1. Obtain medical images of diabetic retinopathy from the hospital;

[0049] Step 2. Preprocess the collected medical images first, and then ask professional ophthalmologists to manually label the lesions in the medical images to form a dataset with labeling information required for training the ResNeSt convolutional neural network classification model. Divide the dataset with labeled information into three parts: training set, validation set and test set;

[0050] Step 3. Build the deep learning server platform required for the experiment, and then write the python code to prepare the model;

[0051] Step 4. By introducing two lightweight and efficient convolution operat...

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 diabetic retinopathy image classification method based on an improved ResNeSt convolutional neural network model. The method comprises the following steps: acquiring a lesion image from a hospital; preprocessing the image, manually marking the image by an ophthalmologist, and dividing a data set; establishing a deep learning server platform required by the experiment then, and then writing a python code; introducing two lightweight and efficient convolution operations of OctConv and SPConv into a ResNeSt convolutional neural network, and introducing a learning rate mediation mechanism of Warm Restart and cosine annealing; pre-training the improved ResNeSt network by adopting an ILSVRC2012 data set, and migrating the obtained model to the preprocessed data set for fine tuning; and loading a test set, testing the trained ResNeSt convolutional neural network classification model to obtain a classification result, and judging whether each classification index meets the requirement or not. The diabetic retinopathy image classification method is realized, the improved ResNeSt model is utilized, the operation efficiency and the classification accuracy are high, and the application value is very high.

Description

technical field [0001] The invention belongs to the technical field of medical image processing, and in particular relates to a diabetic retinopathy image classification method based on an improved ResNeSt convolutional neural network model. Background technique [0002] Diabetes is a body metabolic disease with a relatively high incidence rate, mostly in older groups. It will not only have a serious impact on the normal blood sugar metabolism of the human body, but also cause damage to other parts of the body. Among them, diabetic retinopathy ( DR) is a typical complication of diabetes. DR will not only have a great impact on people's vision, but also cause damage to the brain nerves. DR generally has a latent period of 10-15 years from early stage to blindness, and the early stage is not prominent, so it often does not attract people's attention. [0003] Diabetic retinopathy not only causes blindness to patients, but also causes serious mental and economic burdens to pa...

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
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/241
Inventor 蓝如师焦志勇罗笑南刘振丙汪华登潘细朋
Owner GUILIN UNIV OF ELECTRONIC 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