U-Net-ASPP: U-Net based on atrous spatial pyramid pooling model for medical image segmentation in COVID-19

ABSTRACT

In this paper, we propose a novel U-Net model based on atrous spatial pyramid pooling for medical image segmentation. In the encoder, a pre-trained ImageNet Efficientnet-B0 network is used to extract features for effective information. Atrous spatial pyramid pooling (ASPP) is used to extract the above multi-scale information between encoding and decoding. Finally, the captured information is cascaded with the information of the encoding layer, and the segmentation accuracy is improved by combining the attention mechanism. Experimental results on public data sets show that the accuracy, recall rate and Dice coefficient of the proposed algorithm are 85.35