Medical Image Segmentation Using Squeeze-and-Expansion Transformers

TL;DR

Segtran introduces a transformer-based framework for medical image segmentation that captures large contextual information and maintains high spatial resolution, outperforming traditional multi-scale methods across various 2D and 3D tasks.

Contribution

This work presents Segtran, a novel transformer architecture with a Squeeze-and-Expansion attention mechanism and a new positional encoding, enhancing segmentation accuracy and generalization in medical imaging.

Findings

Achieved highest accuracy on multiple medical segmentation benchmarks.

Demonstrated strong cross-domain generalization capabilities.

Outperformed existing methods in 2D and 3D segmentation tasks.

Abstract

Medical image segmentation is important for computer-aided diagnosis. Good segmentation demands the model to see the big picture and fine details simultaneously, i.e., to learn image features that incorporate large context while keep high spatial resolutions. To approach this goal, the most widely used methods -- U-Net and variants, extract and fuse multi-scale features. However, the fused features still have small "effective receptive fields" with a focus on local image cues, limiting their performance. In this work, we propose Segtran, an alternative segmentation framework based on transformers, which have unlimited "effective receptive fields" even at high feature resolutions. The core of Segtran is a novel Squeeze-and-Expansion transformer: a squeezed attention block regularizes the self attention of transformers, and an expansion block learns diversified representations. Additionally, we propose a new positional encoding scheme for transformers, imposing a continuity inductive bias for images. Experiments were performed on 2D and 3D medical image segmentation tasks: optic disc/cup segmentation in fundus images (REFUGE'20 challenge), polyp segmentation in colonoscopy images, and brain tumor segmentation in MRI scans (BraTS'19 challenge). Compared with representative existing methods, Segtran consistently achieved the highest segmentation accuracy, and exhibited good cross-domain generalization capabilities. The source code of Segtran is released at https://github.com/askerlee/segtran.