Landmark-Free Preoperative-to-Intraoperative Registration in Laparoscopic Liver Resection

2025-04-21

AI Summary

In laparoscopic liver resection, existing preoperative-intraoperative registration methods rely on ambiguous anatomical landmarks and neglect intraoperative deformation modeling, leading to inaccurate spatial localization. This paper proposes a markerless, anatomy-agnostic 3D–3D registration framework that reformulates the conventional 3D–2D pipeline into a two-stage rigid + non-rigid optimization. First, a feature-decoupled Transformer is designed to learn robust cross-modal correspondences. Second, a structural regularization deformable network is introduced, incorporating a low-rank geometric similarity constraint to enhance surface consistency. Leveraging self-supervised learning and a newly constructed structured endoscopic video dataset—P2I-LReg (21 patients, 346 keyframes)—our method significantly outperforms state-of-the-art approaches on both synthetic and real-world data. A clinical user study validates its practical utility and improved localization accuracy.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Liver registration by overlaying preoperative 3D models onto intraoperative 2D frames can assist surgeons in perceiving the spatial anatomy of the liver clearly for a higher surgical success rate. Existing registration methods rely heavily on anatomical landmark-based workflows, which encounter two major limitations: 1) ambiguous landmark definitions fail to provide efficient markers for registration; 2) insufficient integration of intraoperative liver visual information in shape deformation modeling. To address these challenges, in this paper, we propose a landmark-free preoperative-to-intraoperative registration framework utilizing effective self-supervised learning, termed ourmodel. This framework transforms the conventional 3D-2D workflow into a 3D-3D registration pipeline, which is then decoupled into rigid and non-rigid registration subtasks. ourmodel~first introduces a feature-disentangled transformer to learn robust correspondences for recovering rigid transformations. Further, a structure-regularized deformation network is designed to adjust the preoperative model to align with the intraoperative liver surface. This network captures structural correlations through geometry similarity modeling in a low-rank transformer network. To facilitate the validation of the registration performance, we also construct an in-vivo registration dataset containing liver resection videos of 21 patients, called emph{P2I-LReg}, which contains 346 keyframes that provide a global view of the liver together with liver mask annotations and calibrated camera intrinsic parameters. Extensive experiments and user studies on both synthetic and in-vivo datasets demonstrate the superiority and potential clinical applicability of our method.

Problem

• Eliminates reliance on ambiguous anatomical landmarks for liver registration
• Improves intraoperative liver shape deformation modeling accuracy
• Transforms 3D-2D registration into decoupled 3D-3D rigid and non-rigid tasks

Innovation

• Landmark-free registration with self-supervised learning
• Feature-disentangled transformer for rigid transformations
• Structure-regularized deformation network for alignment