Colorectal Cancer Detection and Diagnosis
Colorectal cancer is one of the main causes of cancer death worldwide. The detection and diagnosis of its precursor lesion, the polyp, is key to reduce its prevalence; efforts have been made from the clinical side to ensure a thorough exploration of the colon but still some lesions are missed.
During the last 10 years CVC researchers have been working on the development and validation of several computational methods to assist clinicians in lesion detection and diagnosis. Thanks to the collaboration with Hospital Clinic of Barcelona and the advent of new technologies such as Deep learning, we are now able to tackle both tasks under real time conditions, compulsory in order our methods to be used in the exploration room.
Apart from method development, our team has a strong focus on the definition of validation frameworks, being pioneers in the creation and publication of several fully annotated datasets and annotation platforms that have been already used in several international challenges held in main medical imaging conferences such as ISBI or MICCAI.