Search papers, labs, and topics across Lattice.
This paper introduces a novel 3D-2D liver registration pipeline for surgical augmented reality that leverages laparoscopic depth maps and a foundation pose estimator for initial camera-liver pose estimation. They replace computationally expensive finite-element (FE) based deformation models with non-rigid iterative closest point (NICP) to simplify the registration process. Experiments on real patient data demonstrate that the depth-augmented foundation pose approach, combined with NICP, achieves a mean registration error of 9.91 mm, showcasing its potential as a clinically relevant and engineering-friendly alternative to FE-based methods.
Ditch the complex finite-element models: NICP offers a lightweight, accurate, and engineering-friendly alternative for deformable liver registration in surgical AR.
Augmented reality can improve tumor localization in laparoscopic liver surgery. Existing registration pipelines typically depend on organ contours; deformable (non-rigid) alignment is often handled with finite-element (FE) models coupled to dimensionality-reduction or machine-learning components. We integrate laparoscopic depth maps with a foundation pose estimator for camera-liver pose estimation and replace FE-based deformation with non-rigid iterative closest point (NICP) to lower engineering/modeling complexity and expertise requirements. On real patient data, the depth-augmented foundation pose approach achieved 9.91 mm mean registration error in 3 cases. Combined rigid-NICP registration outperformed rigid-only registration, demonstrating NICP as an efficient substitute for finite-element deformable models. This pipeline achieves clinically relevant accuracy while offering a lightweight, engineering-friendly alternative to FE-based deformation.
