NeuralBoneReg: A Novel Self-Supervised Method for Robust and Accurate Multi-Modal Bone Surface Registration

2025-11-18

AI Summary

In computer-assisted orthopedic surgery, registration between preoperative medical images and intraoperative point clouds is challenging due to modality heterogeneity. To address this, we propose NeuralBoneReg, a self-supervised framework that pioneers the integration of implicit neural unsigned distance fields (UDFs) with self-supervised learning for modality-agnostic bone surface registration—without requiring cross-subject annotations. Our method employs 3D point clouds as a unified representation; an MLP implicitly models preoperative bone geometry and generates transformation hypotheses, enabling sequential global initialization followed by local refinement. NeuralBoneReg achieves robust cross-modal and cross-anatomical registration, substantially enhancing clinical applicability. Evaluated on UltraBones100k, UltraBones-Hip, and SpineDepth datasets, it achieves mean rotation/translation errors of 1.68°/1.86 mm, 1.88°/1.89 mm, and 3.79°/2.45 mm, respectively—outperforming state-of-the-art methods.

Technology Category

• Computer Vision: 3D Computer Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

In computer- and robot-assisted orthopedic surgery (CAOS), patient-specific surgical plans derived from preoperative imaging define target locations and implant trajectories. During surgery, these plans must be accurately transferred, relying on precise cross-registration between preoperative and intraoperative data. However, substantial modality heterogeneity across imaging modalities makes this registration challenging and error-prone. Robust, automatic, and modality-agnostic bone surface registration is therefore clinically important. We propose NeuralBoneReg, a self-supervised, surface-based framework that registers bone surfaces using 3D point clouds as a modality-agnostic representation. NeuralBoneReg includes two modules: an implicit neural unsigned distance field (UDF) that learns the preoperative bone model, and an MLP-based registration module that performs global initialization and local refinement by generating transformation hypotheses to align the intraoperative point cloud with the neural UDF. Unlike SOTA supervised methods, NeuralBoneReg operates in a self-supervised manner, without requiring inter-subject training data. We evaluated NeuralBoneReg against baseline methods on two publicly available multi-modal datasets: a CT-ultrasound dataset of the fibula and tibia (UltraBones100k) and a CT-RGB-D dataset of spinal vertebrae (SpineDepth). The evaluation also includes a newly introduced CT--ultrasound dataset of cadaveric subjects containing femur and pelvis (UltraBones-Hip), which will be made publicly available. NeuralBoneReg matches or surpasses existing methods across all datasets, achieving mean RRE/RTE of 1.68°/1.86 mm on UltraBones100k, 1.88°/1.89 mm on UltraBones-Hip, and 3.79°/2.45 mm on SpineDepth. These results demonstrate strong generalizability across anatomies and modalities, providing robust and accurate cross-modal alignment for CAOS.

Problem

• Addresses modality heterogeneity in bone surface registration for orthopedic surgery
• Enables automatic cross-modal alignment between preoperative and intraoperative data
• Provides robust registration without requiring inter-subject training data

Innovation

• Self-supervised framework using modality-agnostic point clouds
• Implicit neural unsigned distance field for bone modeling
• MLP-based registration with global initialization and local refinement