Members of the MIRTHAI Lab celebrate today for their successful publication to the Journal of Computers in Biology and Medicine. Beneath the watchful eye of Dr. Wei Shao; Amarjeet Kumar, Hongxu Jiang, and Dr. Muhammad Imran have been working tirelessly to attain superior results for one of the lab’s current projects. Their paper titled, A Flexible 2.5D Medical Image Segmentation Approach with In-Slice and Cross-Slice Attention is scheduled to be published to the prestigious journal in the coming months. This journal’s impact factor of 7.0 and its top 9% distinction in Computer Science Applications underscore the hard work and perseverance the collaborators have put into this project.
In this paper, they develop and investigate CSA-Net, a flexible 2.5D segmentation model capable of processing 2.5D images with an arbitrary number of slices through an innovative Cross-Slice Attention (CSA) module. Throughout the course of the paper, CSA-Net is evaluated on three 2.5D segmentation tasks: (1) multiclass brain MRI segmentation, (2) binary prostate MRI segmentation, and (3) multi-class prostate MRI segmentation. The results show that CSA-Net was capable of outperforming leading 2D and 2.5D segmentation methods across the three tasks, demonstrating their novel approach’s efficacy and superiority.
The team’s long-standing collaboration has enabled the MIRTHAI lab to make significant advancements in implementing Artificial Intelligence (AI) in medical imaging. CSA-Net is just one of the many contributions from the lab, with many more projects currently underway.
To access the GitHub repository for the project, please click on the link below.