基于编码器空间信息优化的眼底图像分割

In order to solve the low medical efficiency and other problems existing in the current national medical and health system, the development of intelligent medicine and the establishment of intelligent medical system has become an inevitable choice. As the basis of various medical image processing tasks, medical image segmentation is of great significance for the realization of intelligent medicine. However, due to the problems of small datasets, multiple models, low contrast and large noise interference in the task of medical image segmentation, the accuracy of medical image segmentation using deep learning algorithm cannot meet the clinical requirements. Therefore, this paper focused on two mainstream segmentation tasks, namely, retinal vessel segmentation and liver segmentation, modified the structure of UNet++, proposed an optimization method for the feature extraction ability of the encoder. The Transformer structure was introduced into the encoder to supplement the global context information and the ablation experiments and comparative experiments on the DRIVE dataset show that the proposed method can effectively improve the ability of retinal vessel segmentation. The method not only improves the segmentation accuracy of the network, but also lays a foundation for the clinical use of medical image segmentation.