LocBAM: Advancing 3D Patch-Based Image Segmentation by Integrating Location Contex

2026-01-21

AI Summary

Existing patch-based 3D medical image segmentation methods often suffer from inadequate anatomical context modeling due to the neglect of spatial positional information of local patches within the global volume, thereby compromising segmentation performance. To address this limitation, this work proposes LocBAM—a novel position-aware attention mechanism that explicitly models spatial positional context in 3D patch-based segmentation for the first time, outperforming conventional approaches such as CoordConv. The proposed mechanism integrates seamlessly into patch-based deep segmentation frameworks and demonstrates consistently stable training and significant performance gains across multiple benchmarks, including BTCV, AMOS22, and KiTS23. Notably, LocBAM exhibits pronounced advantages in scenarios with low organ coverage, where contextual cues are especially critical for accurate segmentation.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multimodal Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Patch-based methods are widely used in 3D medical image segmentation to address memory constraints in processing high-resolution volumetric data. However, these approaches often neglect the patch's location within the global volume, which can limit segmentation performance when anatomical context is important. In this paper, we investigate the role of location context in patch-based 3D segmentation and propose a novel attention mechanism, LocBAM, that explicitly processes spatial information. Experiments on BTCV, AMOS22, and KiTS23 demonstrate that incorporating location context stabilizes training and improves segmentation performance, particularly under low patch-to-volume coverage where global context is missing. Furthermore, LocBAM consistently outperforms classical coordinate encoding via CoordConv. Code is publicly available at https://github.com/compai-lab/2026-ISBI-hooft

Problem

• 3D medical image segmentation
• patch-based methods
• location context
• spatial information
• anatomical context

Innovation

• LocBAM
• location context
• 3D medical image segmentation
• patch-based segmentation
• spatial attention