Development of AI tools to assist the treatment of colorectal cancer

Colorectal cancer is the second most frequent cancer in the general population. Accurate detection of precursor lesions, adenomas, is crucial for prevention of cancer related mortality. The detection and certainly the characterisation (i.e. differentiation between benign and malignant polyps) can be challenging with highly variable detection rates between different operators. In the past all lesions needed operation, but early lesions can be treated endoscopically with a local resection. It is however hard to predict how deep a lesion invades in the wall of the large intestine. The aim of this PhD project is to develop a deep learning based AI tool to aid the physician in 1) immediate decision making for the appropriate treatment and 2) predict histology of the tumor.