MeiBRD: Meta-Learning Intraoperative Biomechanical Residual Deformation
For surgeons needing accurate liver registration during surgery, this method improves upon existing biomechanical and data-driven approaches by combining their strengths, though it is incremental in nature.
The paper tackles intraoperative liver registration under sparse measurements and large deformations. The proposed hybrid framework, MeiBRD, learns a residual deformation to correct biomechanical predictions using graph neural diffusion and meta-learning, achieving improved accuracy and generalization over baselines, especially for out-of-distribution cases.
Accurate intraoperative liver registration is challenging due to substantial soft-tissue deformation yet sparse intraoperative measurements. Biomechanical models regularize this ill-posedness with prior knowledge but exhibit persistent prediction bias due to simplifying assumptions, while data-driven learning solutions struggle with data efficiency, generalization, and physical plausibility. We propose a hybrid registration framework that adapts a biomechanical prior using sparse intraoperative correspondences. Rather than learning a full deformation field, we learn a residual deformation function that corrects linear biomechanical predictions, modeled as a graph neural diffusion function with geometry-aware attention over the 3D liver mesh. To enable long-range information transfer of sparse observations, we take a novel perspective of sparse intraoperative measurements as \textit{context} samples where input-output pairs of the residual deformation function are fully observed, casting the problem into learning-to-learn this residual function from intraoperative context samples with feedforward meta-learners. Experiments on a deformable liver phantom dataset demonstrate improved registration accuracy and generalization compared to rigid, biomechanical, and data-driven baselines, particularly for out-of-distribution geometries and deformations.