Improving 3-D Medical Image Segmentation at Boundary Regions Using Local Self-Attention and Global Volume Mixing

Volumetric medical image segmentation is a fundamental problem in medical image analysis where the objective is to accurately classify a given 3-D volumetric medical image with voxel-level precision. In this work, we propose a novel hierarchical encoder–decoder-based framework that strives to explicitly capture the local and global dependencies for volumetric 3-D medical image segmentation. The proposed framework exploits local volume-based self-attention to encode the local dependencies at high resolution and introduces a novel volumetric multi-layer perceptron (MLP)-mixer to capture the global dependencies at low-resolution feature representations, respectively. The proposed volumetric MLP-mixer learns better associations among volumetric feature representations. These explicit local and global feature representations contribute to better learning of the shape-boundary characteristics of the organs. Extensive experiments on three different datasets reveal that the proposed method achieves favorable performance compared to state-of-the-art approaches. On the challenging Synapse Multiorgan dataset, the proposed method achieves an absolute 3.82% gain over the state-of-the-art approaches in terms of HD95 evaluation metrics while a similar improvement pattern is exhibited in Medical Segmentation Decathlon (MSD) Liver and Pancreas tumor datasets. We also provide a detailed comparison between recent architectural design choices in the 2-D computer vision literature by adapting them for the problem of 3-D medical image segmentation. Finally, our experiments on the ZebraFish 3-D cell membrane dataset having limited training data demonstrate the superior transfer learning capabilities of the proposed vMixer model on the challenging 3-D cell instance segmentation task, where accurate boundary prediction plays a vital role in distinguishing individual cell instances.