Technology Category
Computer Vision: SegmentationMachine Learning: Hardware-aware MLMultiagent Systems: Other Foundations of Multi Agent Systems
XAG-Net: A Cross-Slice Attention and Skip Gating Network for 2.5D Femur MRI Segmentation
🤖 AI Summary
Accurate femoral segmentation from MRI is critical for orthopedic diagnosis and treatment, yet existing 2D and 3D deep learning approaches struggle to balance segmentation accuracy and computational efficiency. To address this, we propose XAG-Net—a novel 2.5D U-Net architecture featuring two key innovations: (1) pixel-wise Cross-Slice Attention (CSA), the first mechanism to model fine-grained contextual dependencies among adjacent slices at each spatial location; and (2) an Attention-Gated (AG) skip-connection module that dynamically modulates multi-scale feature flow. Together, CSA and AG jointly enhance inter-slice contextual modeling and intra-slice feature optimization. Extensive experiments on femoral MRI segmentation demonstrate that XAG-Net significantly outperforms standard 2D, 2.5D, and 3D U-Net baselines, achieving a +2.1% improvement in Dice score while maintaining high computational efficiency. Ablation studies confirm the complementary and synergistic contributions of CSA and AG.
Technology Category
Application Category
📝 Abstract
Accurate segmentation of femur structures from Magnetic Resonance Imaging (MRI) is critical for orthopedic diagnosis and surgical planning but remains challenging due to the limitations of existing 2D and 3D deep learning-based segmentation approaches. In this study, we propose XAG-Net, a novel 2.5D U-Net-based architecture that incorporates pixel-wise cross-slice attention (CSA) and skip attention gating (AG) mechanisms to enhance inter-slice contextual modeling and intra-slice feature refinement. Unlike previous CSA-based models, XAG-Net applies pixel-wise softmax attention across adjacent slices at each spatial location for fine-grained inter-slice modeling. Extensive evaluations demonstrate that XAG-Net surpasses baseline 2D, 2.5D, and 3D U-Net models in femur segmentation accuracy while maintaining computational efficiency. Ablation studies further validate the critical role of the CSA and AG modules, establishing XAG-Net as a promising framework for efficient and accurate femur MRI segmentation.
Problem
Research questions and friction points this paper is trying to address.
Accurate femur MRI segmentation for orthopedic diagnosis
Overcoming limitations of 2D and 3D segmentation approaches
Enhancing inter-slice and intra-slice feature modeling
Innovation
Methods, ideas, or system contributions that make the work stand out.
2.5D U-Net with cross-slice attention
Pixel-wise softmax for inter-slice modeling
Skip attention gating for feature refinement
🔎 Similar Papers
No similar papers found.
