PhD Defence Mark Witteveen | Shedding Light on Colorectal Cancer: Advanced Imaging Technology To Guide Surgeons During Keyhole Surgery

Shedding Light on Colorectal Cancer: Advanced Imaging Technology To Guide Surgeons During Keyhole Surgery

Mark Witteveen is a PhD student in the Department of Nanobiophysics. (Co)Promotors are prof.dr. T.J.M. Ruers and prof.dr.ir. R.M. Verdaasdonk from the Faculty of Science & Technology and prof. dr. H.J.C.M. Sterenborg from the Erasmus University Rotterdam.

Colorectal cancer remains one of the most common and deadly forms of cancer worldwide. While surgeons increasingly treat patients using minimally invasive keyhole surgeries to speed up recovery, operating through tiny incisions severely restricts their field of view. This makes it challenging to accurately locate tumors, especially when they are hidden just beneath the tissue surface.

This dissertation explores a powerful solution: hyperspectral imaging (HSI). This advanced camera technology utilizes broadband light to capture how different wavelengths reflect off tissue. Because healthy and cancerous tissues have distinct biochemical and structural properties, they reflect light differently. By decoding these optical signatures, the camera can act as a real-time visual guide for the surgeon.

To bring this technology closer to the operating room, a hardware prototype was developed, extending standard imaging into the invisible short-wave infrared spectrum. To overcome the slow processing speeds and complex calibration that usually plague these systems, a physics-inspired artificial intelligence (AI) network was designed. This AI processes tissue data 3,000 times faster than traditional methods, paving the way for instant feedback during surgery.

The clinical relevance of the device was successfully validated on a dataset of over forty colorectal cancer specimens. Using weak-label machine learning, the system reliably localized the tumors. This work establishes a solid foundation for the future integration of smart optical guidance into computer-assisted and robotic surgery.