Research on Medical Image Segmentation Method Based on Semi Supervised Learning

Medical image segmentation is a key task in medical image analysis, which is widely used in tumor detection, organ segmentation and other fields. However, the annotation of medical image data is costly and time-consuming, which limits the application of traditional supervised learning methods. In this paper, we first introduce the concepts of image segmentation and semi-supervised learning. Secondly, the FPS-Net based on semi-supervised learning is designed. The network combines the feature pyramid module (FPM) and the cross-attention module (CSA) to improve the segmentation accuracy and computational efficiency. Finally, through experiments on liver CT image dataset and skin lesion dataset, the effectiveness of the proposed method in the semi-supervised learning environment is verified, and the performance of the segmentation model is further improved.