MIT CSAIL's xvr brings patient-specific neural networks to surgical navigation in Nature
EricTopol · x · 2026-09-16
- MIT CSAIL and collaborators published xvr in Nature: a self-supervised framework combining patient-specific neural networks with gradient-based optimization for automatic 2D/3D registration of preoperative CT/MRI volumes to intraoperative X-ray fluoroscopy.
- Existing methods either need per-patient hyperparameter tuning or large labeled datasets limited to trained anatomy.
- xvr is label-free and generalizes across fluoroscopy-guided procedures, improving navigation for image-guided interventions and surgical robotics.
Related event: MIT's xvr Aligns 2D X-rays with 3D Scans in Seconds, Published in Nature(2 posts)→
More from Research
- New edition of Christopher Amato's intro book on cooperative multi-agent RL now available — dhadfieldmenell · 2026-09-17
- RCT finds LLMs fail to significantly boost novices' end-to-end wet-lab skills — anshulkundaje · 2026-09-17
- Noah Giansiracusa guest posts on Terry Tao's blog: let the diners into the kitchen — AlexKontorovich · 2026-09-17
- ICL emerges across language, genomes, proteins, and more — the Convergent Emergence Hypothesis — DanielKhashabi · 2026-09-17
- In Dec 2024, AI Researchers Didn't Expect a Millennium Problem Solved Until 2054 — Tolopono · 2026-09-17
- UChicago researcher hiring predoc for AI x econ with co-first-author papers — ethayarajh · 2026-09-17