Medical image enhancement method and device, equipment and medium

A medical imaging and image enhancement technology, applied in the computer field, can solve the problems of discriminators and classifiers that are difficult to train in a balanced state, poor stability, etc., to improve image enhancement quality, effective extraction and enhancement, and improve stability Effect

Active Publication Date: 2021-09-21
HUAQIAO UNIVERSITY
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, during the training process, it is difficult to train the discriminator and the classifier to a balanced state, and the stability during the training process is not good.

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
  • Medical image enhancement method and device, equipment and medium
  • Medical image enhancement method and device, equipment and medium
  • Medical image enhancement method and device, equipment and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] This embodiment provides a method for enhancing medical images, such as figure 1 shown, including;

[0040] S1. Obtain the input feature map of the medical image to be processed as a data set, randomly select some images from the data set as a training set, and construct an improved ACGAN model.

[0041] Such as figure 2 As shown, the ACGAN framework can use auxiliary classification labels to generate higher-quality samples, and combine the discriminator with the classifier, so that the improved discriminator can not only identify true and false data, but also distinguish different categories of data. The improved ACGAN model of the present invention uses the difference between the Wasserstein distance meter generated data and the real data on the basis of the ACGAN framework to construct an improved ACGAN. A gradient penalty (GP) is used to replace the weight clipping to realize the K-Lipschitz condition and improve the stability of the network training process. Th...

Embodiment 2

[0060] Such as Figure 4 As shown, a kind of medical image enhancement device is provided in this embodiment, comprising:

[0061] The image preparation module is used to obtain the input feature map of the medical image to be processed as a data set, and randomly select some images from the data set as a training set,

[0062] The ACGAN improved model building block is used to build the ACGAN improved model, and the resulting ACGAN improved model uses the Wasserstein distance to calculate the difference between the generated data and the real data for true and false discrimination, and uses gradient penalties instead of weight clipping to achieve K- Lipschitz condition, adding a gradient penalty term L to the loss function gp , so that the loss function of the overall training is:

[0063]

[0064] Equation 2, L s In order to record the loss of true and false judgment of data, L c To record the loss of data classification, D stands for discriminator, G stands for gener...

Embodiment 3

[0076] This embodiment provides an electronic device, such as Figure 5 As shown, it includes a memory, a processor, and a computer program stored in the memory and operable on the processor. When the processor executes the computer program, any implementation manner in Embodiment 1 can be realized.

[0077] Since the electronic device introduced in this embodiment is the device used to implement the method in Embodiment 1 of this application, based on the method described in Embodiment 1 of this application, those skilled in the art can understand the electronic device of this embodiment. Specific implementation methods and various variations thereof, so how the electronic device implements the method in the embodiment of the present application will not be described in detail here. As long as a person skilled in the art implements the equipment used by the method in the embodiment of the present application, it all belongs to the protection scope of the present application. ...

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 medical image enhancement method and device, equipment and a medium, and the method comprises the steps of obtaining an input feature map of a to-be-processed medical image as a data set, and constructing an ACGAN improved model; calculating the difference between the generated data and the real data by using a Wasserstein distance to carry out true and false discrimination, replacing weight cutting by using gradient penalty to realize a K-Lipschitz condition, and adding a gradient penalty term into the loss function; after the ACGAN improved model is trained, inputting a data set, outputting a synthesized image through a generator, and adding the synthesized image into the data set for data enhancement to obtain an enhanced data set; taking a discriminator as a sample feature extractor, and performing feature extraction on the enhanced data set to obtain a sample feature map; and fusing with the input feature map to generate an enhanced image of the to-be-processed medical image. Therefore, the problem that details of the medical image are fuzzy due to overexposure or insufficient exposure is solved.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a medical image enhancement method, device, equipment and medium. Background technique [0002] Medical imaging is a technology to assist doctors in diagnosis, but the quality of images captured by different imaging equipment is uneven. The clarity of some images is low, and from the image, some parts of the tissue are in a connected state, all of which cause great difficulties for image diagnosis. Moreover, the number of cases handled by grassroots doctors is limited, lack of training, and it is difficult to have a high level of diagnosis. Therefore, research on computer-aided diagnosis algorithms for medical imaging diagnosis has been going on in recent years. [0003] Image enhancement technology is one of the key technologies in the study of computer-aided diagnosis algorithms. This technology is used to improve and enhance the quality of the original image, and e...

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): G06T5/50G06T5/00G06K9/62
CPCG06T5/50G06T5/001G06F18/214Y02T10/40
Inventor 郑力新郭铮铮严潭苏秋玲
Owner HUAQIAO UNIVERSITY
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