Improving cancer surgery by developing a smart laparoscope

In this project the Netherlands Cancer Institute, the Amsterdam University Medical Centers, Stryker and Offroad Medical will join forces to develop a new laparoscope combining hyperspectral imaging with state of the art machine learning to identify cancer tissue during surgery.  

In up to 20-30% of cancer patients treated with surgery, either tumor tissue is left behind or critical healthy structures, like nerves, are damaged leading to long lasting complications. Laparoscopic procedures are becoming the preferred technique for many types of cancer surgery, e.g. in the Netherlands 80% of all colorectal cancer surgeries are performed laparoscopically. A smart laparoscope that can identify tumor tissue and critical healthy structures during surgery thus has the potential to make a big impact on cancer patients. 

In regular laparoscopy a camera is attached to the laparoscope that creates an image that is interpreted by the surgeon. In this project, we will improve this concept in two ways. First of all, we will develop a hyperspectral laparoscope, which images tissue from the visible to the near-infrared wavelength range – far beyond what is visible by eye. Secondly, we will use state-of-the-art machine learning to develop software that interprets these hyperspectral images and provides the surgeon with a tissue type for each pixel within the image.   

At the end of the project we will deliver a prototype of the hyperspectral laparoscope and the tissue classification software. Furthermore, we will demonstrate the clinical feasibility of the hyperspectral laparoscopy system by performing in vivo measurements in a small group of colorectal cancer patients.