Cross-modal medical image segmentation method based on symmetric adaptive network

Abstract

The invention discloses a cross-modal medical image segmentation method based on a symmetric adaptive network, and the method comprises the steps: carrying out the preprocessing of a pre-obtained medical image, and obtaining a source domain data set and a target domain data set; constructing a symmetric adaptive network: adopting two symmetric conversion sub-networks which share an encoder to generate a cross-domain image, and mining rich semantic information by using images of different styles; performing optimization training on the symmetric adaptive network based on the source domain data set and the target domain data set; and testing the target image based on the optimized and trained symmetric adaptive network to obtain a final medical image segmentation result. According to the method, the distribution difference between a source domain and a target domain is reduced in two aspects of two-way close feature distribution by using two symmetrical conversion sub-networks and mining rich semantic information by using images of different styles; and therefore, good segmentation performance can be obtained on the target domain image, and high practical value can be realized.

Description

technical field [0001] The invention belongs to the field of image segmentation, in particular to a cross-modal medical image segmentation method based on a symmetrical adaptive network. Background technique [0002] In recent years, deep convolutional neural network methods have achieved major breakthroughs in medical image segmentation tasks. Most segmentation tasks often assume that the training set and test set images come from the same data distribution, but in real scenarios, especially in the medical field, due to different acquisition parameters or imaging methods, the training set and test set images often have larger distribution differences. This distribution difference often leads to a sharp drop in the performance of the trained model on the test images. [0003] To alleviate the above problems, a relatively straightforward way is to fine-tune the trained source-domain model with labeled target-domain images. However, annotating pixel-level labels for target ...

AI Technical Summary

Problems solved by technology

Since the feature space in the segmentation model contains a large number of different feature information and is extremely complex, it is difficult to completely eliminate the distribution difference

Images