UD-Mamba: A pixel-level uncertainty-driven Mamba model for medical image segmentation

TL;DR

UD-Mamba introduces a pixel-level uncertainty-driven scanning mechanism to enhance medical image segmentation by focusing on high-uncertainty regions and improving local feature modeling.

Contribution

The paper proposes a novel uncertainty-driven scanning approach for Mamba models, incorporating sequential and skip scanning techniques with learnable parameters and a cosine consistency loss.

Findings

Improved segmentation accuracy across multiple medical imaging datasets.

Effective modeling of local features and ambiguous boundaries.

Enhanced interaction between background and foreground regions.

Abstract

Recent advancements have highlighted the Mamba framework, a state-space model known for its efficiency in capturing long-range dependencies with linear computational complexity. While Mamba has shown competitive performance in medical image segmentation, it encounters difficulties in modeling local features due to the sporadic nature of traditional location-based scanning methods and the complex, ambiguous boundaries often present in medical images. To overcome these challenges, we propose Uncertainty-Driven Mamba (UD-Mamba), which redefines the pixel-order scanning process by incorporating channel uncertainty into the scanning mechanism. UD-Mamba introduces two key scanning techniques: 1) sequential scanning, which prioritizes regions with high uncertainty by scanning in a row-by-row fashion, and 2) skip scanning, which processes columns vertically, moving from high-to-low or low-to-high uncertainty at fixed intervals. Sequential scanning efficiently clusters high-uncertainty regions, such as boundaries and foreground objects, to improve segmentation precision, while skip scanning enhances the interaction between background and foreground regions, allowing for timely integration of background information to support more accurate foreground inference. Recognizing the advantages of scanning from certain to uncertain areas, we introduce four learnable parameters to balance the importance of features extracted from different scanning methods. Additionally, a cosine consistency loss is employed to mitigate the drawbacks of transitioning between uncertain and certain regions during the scanning process. Our method demonstrates robust segmentation performance, validated across three distinct medical imaging datasets involving pathology, dermatological lesions, and cardiac tasks.