Attention guided matching improves accuracy in medical image alignment
ACorD: Adaptive Correspondence Discovery for Medical Image Registration
Computer Vision and Pattern RecognitionArtificial Intelligence
Summary
Aligning medical images helps doctors compare anatomy across scans, but finding exact matching points is hard and slow. The authors designed a new method called ACorD that looks at multiple possible matches and uses an attentive process to decide which ones are most reliable. This approach reduces errors made early on and improves the alignment step-by-step. Testing on medical image datasets showed that this adaptive technique works better on various body parts than previous methods.
What this means in practice
- •For medical imaging software developers: Enhance image alignment tools by integrating adaptive attention mechanisms to improve registration accuracy across different anatomical regions.
- •For radiology workflow engineers: Improve multi-modal image alignment speed and robustness in clinical workflows ensuring more reliable comparisons of 3D scans.
Authors
Arghavan Rezvani, Kun Han, Xiaohui Xie
Abstract
Medical image registration is a fundamental task for aligning anatomical structures across images. Exhaustively searching for point-wise correspondences is prohibitively expensive in both computation and memory, particularly for high-resolution 3D medical images. Most recent learning-based methods directly predict correspondences in a coarse-to-fine manner, but often propagate intermediate estimates across scales without explicitly weighting correspondence evidence, which can allow coarse-scale errors to persist during refinement. To address these challenges, we propose ACorD, an adaptive correspondence discovery framework designed to mitigate error propagation with an attention-guided process. ACorD first predicts multiple potential correspondence candidates through Correlation-guided Cross-scale Offset Prediction (CCOP), which combines correlation maps as spatial matching priors with cross-scale features as global deformation context. Then, Multi-Head Deformable Aggregation (MHDA) aggregates these candidate correspondences using attention, producing a learned, non-uniform weighting over competing matches rather than treating them equally. The multi-head design further captures complementary deformation patterns, ensuring both global alignment and local precision. By integrating adaptive offset prediction with selective, attention-based aggregation of correspondence evidence, ACorD enables accurate deformation refinement. Extensive experiments on four public medical registration benchmarks demonstrate that adaptive correspondence discovery within a multi-scale refinement framework improves registration robustness across diverse anatomical structures.