Medical image style conversion method and apparatus, medical device, image system and storage medium

Abstract

The invention relates to a medical image style conversion method and apparatus, a medical device, an image system and a storage medium. The method comprises the following steps of: acquiring a neuralnetwork model; acquiring original medical images; using the original medical image as the input of the neural network model to obtain the target medical image by the neural network model, wherein Thedifference between the content features of the original medical image and that of the target medical image is within the preset threshold, and the style features of the original medical image are different from those of the target medical image. According to the above method, through the neural network based on style transfer algorithm, the original medical images are transformed into target medical images with different styles. The invention can be applied to the style conversion between the medical images of various modes, and can keep the change of the content characteristics of the imageswithin the preset threshold value and not easily cause distortion, so that the doctor can conveniently compare and observe the converted target medical images with the medical images of other modes.

Description

technical field [0001] The present invention relates to the field of medical images, in particular to a style conversion method and device for medical images, medical equipment, imaging systems and storage media. Background technique [0002] In medical diagnosis and treatment, it is often necessary to take different modal medical images such as computed tomography, magnetic resonance, and X-rays in the same area for comparative observation. If it is too large, it may make it difficult for the contents of the pictures to correspond to each other, which will cause inconvenience to diagnosis and treatment. [0003] The traditional style feature conversion between different modalities of medical images is generally through the method of grayscale transformation or texture-based mapping method. However, the grayscale range of medical images is unstable, and the transformation methods are different. The transformation of style features through grayscale transformation will easil...

AI Technical Summary

Problems solved by technology

However, the grayscale range of medical images is unstable, and the transformation methods are different. The transformation of style features through grayscale transformation will easily lead to image distortion; and because the information of medical images cannot be described by simple textures, texture mapping The method of transforming style features is relatively poor in adaptability

Images