RMS-UNet: Residual multi-scale UNet for liver and lesion segmentation

作者：

Highlights：

• A novel loss-function to avoid local minima trap for segmentation.

• A Residual UNet with the dilated convolution to probe inter-slice gener-alized features.

• Extensive comparative analysis on four publicly available datasets.

摘要

•A novel loss-function to avoid local minima trap for segmentation.•A Residual UNet with the dilated convolution to probe inter-slice gener-alized features.•Extensive comparative analysis on four publicly available datasets.

论文关键词：Abdominal CT,Liver and lesion segmentation,Deep learning,Residual,Dilated convolution

论文评审过程：Received 17 July 2020, Revised 12 November 2021, Accepted 17 December 2021, Available online 12 January 2022, Version of Record 20 January 2022.

论文官网地址：https://doi.org/10.1016/j.artmed.2021.102231