Medical image synthesis method based on double generative adversarial networks

Abstract

The invention provides a medical image synthesis method based on double generative adversarial networks. The method comprises the main content of performing data management, generating the generativeadversarial networks, training a U-NET, establishing assessment indexes and obtaining processed pictures, wherein first, a DRIVE database is used to manage a first-stage GAN, then the first-stage GANgenerates a partitioning mask representing variable geometry of a dataset, a second-stage GAN converts the mask produced at the first stage into an image with a sense of reality, an generator minimizes a loss function of the true data in classification through a descriminator, then the U-NET is trained to assess the reliability of synthetic data, and finally the assessment indexes are establishedto measure a generated model. According to the method, by use of a pair of generative adversarial networks to create a new image generation path, the problem that the synthetic image contains a fake shadow and noise is solved, the stability and the sense of reality of the image are improved, and meanwhile image details are clearer.

Description

technical field [0001] The invention relates to the field of image synthesis, in particular to a medical image synthesis method based on dual generation confrontation networks. Background technique [0002] With the development of image processing technology, people are currently very interested in the methods of medical image classification. However, medical imaging data is scarce, expensive, and fraught with legal issues regarding patient privacy. Typical patient data consent forms only allow images to be used in medical journals or for educational purposes, which means that most medical data is not available for general public research. Synthesizing medical images was thought of as a solution to these problems. Using the method of medical image synthesis, researchers will have a large amount of medical image data as research materials, and can perform computer-aided medical diagnosis, assist medical professionals to interpret medical images, and classify medical images. ...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem that images generated by GAN may still contain artifacts and noises, resulting in the instability of synthesized images, the purpose of the present invention is to provide a medical image synthesis method based on dual generative confrontation networks, first using the DRIVE database to manage The first stage GAN, then the first stage GAN generates a segmentation mask representing the variable geometry of the dataset, the second stage GAN converts the mask made in the first stage into a photorealistic image, and the generator passes the discriminator to classify Minimize its loss function for real data, then train U-NET to evaluate the reliability of synthetic data, and finally establish an evaluation index to measure the generated model

Images