Is Long Range Sequential Modeling Necessary For Colorectal Tumor Segmentation?

TL;DR

This paper evaluates the necessity of long-range sequence modeling in 3D colorectal tumor segmentation, finding that local interactions can outperform global models in complex, small-region cases, challenging current trends.

Contribution

It introduces a new dataset and demonstrates that local token interactions can surpass long-range models like Transformers in specific segmentation scenarios.

Findings

Local token interactions outperform long-range models in small, complex regions

Proposes a shift in 3D tumor segmentation research focus

Introduces the CTS-204 dataset for colorectal tumor segmentation

Abstract

Segmentation of colorectal cancer (CRC) tumors in 3D medical imaging is both complex and clinically critical, providing vital support for effective radiation therapy planning and survival outcome assessment. Recently, 3D volumetric segmentation architectures incorporating long-range sequence modeling mechanisms, such as Transformers and Mamba, have gained attention for their capacity to achieve high accuracy in 3D medical image segmentation. In this work, we evaluate the effectiveness of these global token modeling techniques by pitting them against our proposed MambaOutUNet within the context of our newly introduced colorectal tumor segmentation dataset (CTS-204). Our findings suggest that robust local token interactions can outperform long-range modeling techniques in cases where the region of interest is small and anatomically complex, proposing a potential shift in 3D tumor segmentation research.