Endoscopic Vision challenge: Automatic polyp detection in colonoscopy videos

October 9, 2015 at 9:00 am by

Colonoscopy is the primary method for colon cancer screening and prevention. However, according to clinical studies, during the colonoscopy procedures, one out of every four polyps is missed.  Patients with missed polyps may develop a late-stage colon cancer with a survival rate of less than 10%. An automatic polyp detection system can point out the locations of polyps in the videos, increasing the attentiveness of colonoscopists during procedures, helping them reduce their polyp miss-rates. Given the large population that annually undergoes colonoscopy, even a modest decrease in polyp miss-rate can save a lot of lives.

Overall, the sub-challenge can be divided in two different tasks, participants are invited to submit results for at least one of the following tasks:

Polyp localization: Can your method cope with high polyp appearance variability and, therefore, accurately indicate the area of the image where the polyp is?

Polyp detection: Can your method state whether there is a polyp or not in the image? Which is the delay between polyps’ first apparition and your first detection?

Important Dates: (Preliminary Timeline)

June, 15th: Release of training data

July, 24th: Release of test data

September, 20th: Submission deadline for results

October, 9th: Challenge day