A flexible 2.5D medical image segmentation approach with in-slice and cross-slice attention

Amarjeet Kumar, Hongxu Jiang, Muhammad Imran, Cyndi Valdes, Gabriela Leon, Dahyun Kang, Parvathi Nataraj, Yuyin Zhou, Michael D. Weiss, Wei Shao

Abstract: Highlights•2D and 3D models face accuracy and efficiency challenges in segmenting 2.5D images.•We proposed CSA-Net, a 2.5D model designed for the segmentation of 2.5D images.•CSA-Net uses cross-slice attention to effectively capture inter-slice correlations.•CSA-Net outperformed leading 2D, 2.5D, and 3D models across three 2.5D datasets.