Graph method improves medical image analysis across slices
G^2RA-NET: Graph-based Cross-Slice Relation Modeling with Attention Gating for Medical Image Segmentation
Computer Vision and Pattern Recognition
Summary
Medical images like MRIs are made of many slices, and understanding how these slices relate is important for accurate diagnosis. Existing methods struggle to connect information between slices, which lowers the quality of detection. This paper presents a new technique called G²RA-Net that treats each slice as a point in a network to share information efficiently, and uses focus methods to highlight important details. Tests on brain and abdomen scans show this approach makes segmentation more precise and consistent.
What this means in practice
- •For medical imaging developers: Build more consistent brain and abdominal image segmentation tools by modeling relationships between neighboring slices effectively.
- •For radiology software engineers: Enhance segmentation accuracy and boundary detail in existing diagnostic software that processes volumetric medical scans.
Authors
Shengye Wang, Zonglin Wu, Liang Fan, Yule Xue, Haozhe Zhao
Abstract
Medical image segmentation supports quantitative clinical analysis and computer-aided diagnosis. Recent methods for medical image segmentation have improved both local feature representation and volumetric context modeling. However, existing methods still strug- gle to efficiently model cross-slice relations in anisotropic volumet- ric images, limiting segmentation consistency and accuracy. This pa- per proposes G^2RA-Net, a medical image segmentation framework that combines graph-based cross-slice relation modeling with atten- tion gating. Graph-Based Slice Relationship Modeling (GSRM) cap- tures anatomical dependencies across consecutive slices by repre- senting each slice as a graph node and propagating semantic con- text through graph message passing. The Cross-Slice Attention Gate (CSAG) then selects relevant neighboring context and emphasizes target anatomical regions through attention-guided feature modula- tion. Experiments on brain MRI and abdominal CT datasets demon- strate that G^2RA-Net outperforms representative methods in seg- mentation accuracy and boundary quality. Ablation studies further validate the proposed design.