NEWS · RESEARCH · #483
MIT researchers develop xvr — an AI that rapidly aligns surgical X‑rays with patients’ 3D scans
Researchers at MIT’s CSAIL and clinical collaborators developed xvr, an AI system that adapts to an individual patient in about five minutes and automatically registers intraoperative X‑rays to preoperative 3D scans in seconds with reported sub‑millimeter precision. The method, published in Nature, outperformed prior AI approaches by roughly an order of magnitude across diverse patients, body parts, and procedures.
KEY POINTS
- Researchers at MIT’s CSAIL and clinical collaborators developed xvr, an AI system that adapts to an individual patient in about five minutes and automatically registers intraoperative X‑rays to preoperative 3D scans in seconds with reported sub‑millimeter precision.
- The method, published in Nature, outperformed prior AI approaches by roughly an order of magnitude across diverse patients, body parts, and procedures.
- By enabling fast, patient‑specific 2D/3D registration with high precision, xvr could make image guidance for minimally invasive procedures faster, safer, and more broadly usable.
WHY IT MATTERS
By enabling fast, patient‑specific 2D/3D registration with high precision, xvr could make image guidance for minimally invasive procedures faster, safer, and more broadly usable.