The IntraSurge project is an innovative research initiative focused on advancing surgical procedures through intelligent, AI-driven guidance systems. Our aim is to enhance intraoperative anatomy recognition, support surgeons in real time, prevent unnecessary tissue injury, and ultimately improve patient safety.
This project is proudly funded by the Dutch Research Council (NWO) and the Hanarth Fonds, and involves close collaboration with esteemed academic institutions including Eindhoven University of Technology, University Medical Center Utrecht (UMCU), and Utrecht University. UMCU is a leading center for robotic-assisted esophagectomy and maintains a comprehensive database of monocular and stereoscopic surgical video footage. We also collaborate with industry partners Rods&Cones and Asensus Surgical.
Our current research interests include:
- Video-based anatomical segmentation
- Surgical phase recognition
- Surgical foundation models
- Vision-language modeling
- 3D-3D registration: Aligning preoperative CT scans with intraoperative video
- Depth estimation & SLAM
- Clinical value validation
These efforts are designed to push the boundaries of computer-aided surgery and deliver tangible benefits to both clinicians and patients.