Method for optimizing medical image classification performance based on generative adversarial network

A medical image and network optimization technology, applied in neural learning methods, biological neural network models, and recognition of medical/anatomical models, can solve problems such as over-resampling of positive sample data, over-fitting of classification models, and insufficient data. Achieve good robustness, reduce collection costs, and avoid overfitting effects

Inactive Publication Date: 2019-04-16
杭州健培科技有限公司
View PDF4 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a method for optimizing the performance of medical image classification based on generative confrontation network, which aims to solve the problem of developing methods using general classification algorithms, which are often prone to overfitting of classification models due to insufficient data or excessive resampling of positive sample data. Together, the problem of poor robustness

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
  • Method for optimizing medical image classification performance based on generative adversarial network
  • Method for optimizing medical image classification performance based on generative adversarial network
  • Method for optimizing medical image classification performance based on generative adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0018] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0019] as attached figure 1 As shown, a method for optimizing medical image classification performance based on generative confrontation network proposed by the present invention, its main steps include: constructing a classification task data set; training classification algorithm model on existing data; using generative confrontation network to generate new Positive sample candidate data; use the voting mechanism to strictly screen the generated positive sample data; integrate the generated data into the existing positive sample data in a certain proportion to fine-tune the classification network.

[0020] The present invention is applicable to the development of classification models for different medical image data. In order to facilitate the understanding of various details in the invention, the development of a classification model for ...

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 method for optimizing medical image classification performance based on a generative adversarial network. The method comprises the steps of constructing a classification taskdata set; training a classification algorithm model on the existing data; generating new positive sample candidate data by using the generative adversarial network; strictly screening the generated positive sample data by utilizing a voting mechanism; and fusing the generated data into the existing positive sample data fine tuning classification network according to a certain proportion. The medical image classification method has the technical advantages that the problems of poor algorithm generalization capability, easy overfitting, high manual data accumulation cost and the like caused by small positive sample data size in medical image classification are solved. In addition, the anti-attack capability of the algorithm can be improved to a certain extent while the performance of the medical image classification algorithm is improved.

Description

technical field [0001] The invention relates to the technical field of image generation and image classification methods, in particular to a method for optimizing the performance of medical image classification based on generative confrontation networks. Background technique [0002] With the development of artificial intelligence technology in recent years, advanced technology has benefited all areas of people's lives. The concept of medical artificial intelligence has also begun to be well known to the public. At present, China's medical system is in an important stage of development. China's medical industry is still facing problems including the uneven distribution of medical resources and the large population base of medical patients. In addition, the continuous development of China's economy will soon realize the background of a well-off society in an all-round way. Under the circumstances, people pay more and more attention to their own health problems. Therefore, m...

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): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V2201/03G06N3/044G06N3/045G06F18/2411G06F18/214
Inventor 夏海琪程国华季红丽
Owner 杭州健培科技有限公司
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