UBone3D improves 3D bone shape from noisy ultrasound data
UBone3D: Physics-Rectified Conditional Flow Matching for Anatomical 3D Shape Completion from Ultrasound
Computer Vision and Pattern Recognition
Summary
Ultrasound scans can create partial, unclear 3D pictures of bones because of artifacts like blurring and missing parts. The authors designed UBone3D, a computer method that fills in missing bone shapes by learning from clearer CT scans and simulating how ultrasound imaging causes artifacts. This approach helps produce more accurate and realistic 3D bone models from ultrasound data alone. It was tested on both simulated and real patient data and showed better results than previous methods.
What this means in practice
- •For medical imaging developers: Create accurate 3D bone reconstructions from partial and noisy ultrasound scans for safer, radiation-free patient monitoring.
- •For medical device engineers: Integrate physics-aware shape completion into ultrasound devices to improve real-time anatomical visualization during clinical exams.
Authors
Weiying Chen, Yuchong Gao, Siyuan Li, Marek Reformat, Rui Zheng, Edmond Lou
Abstract
Three-dimensional ultrasound (US) is a safe, radiation-free complementary modality to CT and X-rays for longitudinal monitoring, yet its segmentation-derived partial point clouds are extremely artifact-laden. Consequently, it is challenging to recover a clean and complete anatomical structure from such US point clouds. In this paper, we present UBone3D, a novel framework based on physics-rectified conditional flow matching (CFM) that performs point cloud completion directly from partial US observations. UBone3D models deterministic physics artifacts (e.g., surface thickening, streaking, dropouts) via a simulated physics proxy, and introduces test-time physics rectification to steer the shape completion. At inference, the completion is jointly steered by two decoupled forces: (1) anatomical plausibility enforced by a CT-trained generative shape prior, BoneFM, and (2) physics consistency enforced by USimNet in the ultrasound formation space. Extensive experiments on simulated and in-vivo data demonstrate significant improvements in reconstruction accuracy and anatomical fidelity over existing baselines.