English

Intraoperative 2D/3D Image Registration via Differentiable X-ray Rendering

Computer Vision and Pattern Recognition 2025-03-21 v2

Abstract

Surgical decisions are informed by aligning rapid portable 2D intraoperative images (e.g., X-rays) to a high-fidelity 3D preoperative reference scan (e.g., CT). 2D/3D image registration often fails in practice: conventional optimization methods are prohibitively slow and susceptible to local minima, while neural networks trained on small datasets fail on new patients or require impractical landmark supervision. We present DiffPose, a self-supervised approach that leverages patient-specific simulation and differentiable physics-based rendering to achieve accurate 2D/3D registration without relying on manually labeled data. Preoperatively, a CNN is trained to regress the pose of a randomly oriented synthetic X-ray rendered from the preoperative CT. The CNN then initializes rapid intraoperative test-time optimization that uses the differentiable X-ray renderer to refine the solution. Our work further proposes several geometrically principled methods for sampling camera poses from SE(3)\mathbf{SE}(3), for sparse differentiable rendering, and for driving registration in the tangent space se(3)\mathfrak{se}(3) with geodesic and multiscale locality-sensitive losses. DiffPose achieves sub-millimeter accuracy across surgical datasets at intraoperative speeds, improving upon existing unsupervised methods by an order of magnitude and even outperforming supervised baselines. Our code is available at https://github.com/eigenvivek/DiffPose.

Keywords

Cite

@article{arxiv.2312.06358,
  title  = {Intraoperative 2D/3D Image Registration via Differentiable X-ray Rendering},
  author = {Vivek Gopalakrishnan and Neel Dey and Polina Golland},
  journal= {arXiv preprint arXiv:2312.06358},
  year   = {2025}
}

Comments

CVPR 2024

R2 v1 2026-06-28T13:47:04.142Z