Liver segmentation in abdominal CT images via auto-context neural network and self-supervised contour attention

作者：

Highlights：

• Auto-context algorithm for CNN-based medical image segmentation.

• Self-supervised learning for effective contour attention.

• Superior performance on generalizing neural network.

• Introduction of N-fold cross-validation metric to analyze generalization performance.

摘要

•Auto-context algorithm for CNN-based medical image segmentation.•Self-supervised learning for effective contour attention.•Superior performance on generalizing neural network.•Introduction of N-fold cross-validation metric to analyze generalization performance.

论文关键词：Auto-context neural network,Contour attention network,High-level residual shape prior,Liver segmentation,Self-supervised neural network

论文评审过程：Received 10 May 2020, Revised 13 November 2020, Accepted 18 January 2021, Available online 23 January 2021, Version of Record 2 February 2021.

论文官网地址：https://doi.org/10.1016/j.artmed.2021.102023