Segment Medical Image Using U-Net Combining Recurrent Residuals and Attention

摘要

Medical image segmentation is the key to decide the issue of medical images in clinical practice that can provide a reliable basis. The development of medical image segmentation technology not only affects the development of other related technologies in medical image processing, such as visualization 3D reconstruction, but in the analysis of biomedical images also occupies an extremely important position. With the application of deep learning algorithms in medical image segmentation, medical image segmentation technology has made significant progress. In this paper, we discuss the segmentation method of 2D medical images about U-net variant network. Use the U-net combing recurrent residual model and attention model to segmented the image can get better result.