| Challenge | Dates | Submission | Organizers | Venue |
Ulcerative colitis (UC) is a chronic intestinal inflammatory disease often developed in the colon and rectum. Colorectal cancer caused by UC causes about 1.8 million deaths each year, accounting for more than 12% of global mortality. As one of the early manifestations of cancer, UC should be diagnosed, monitored, and treated early. Colonoscopy is a gold standard clinical procedure widely used for early screening of UC. Scoring of UC can be extended to several different scoring protocols, of which the Mayo Endoscopic Score (MES) is considered one of the most widely used assessments to measure UC activity. MES will be characterized and graded according to vascular pattern, mucosal topology, structural rectal bleeding, etc. However, the grading and interpretation of UC in colonoscopy highly depends on the clinician's experience level. Although experts can assign follow-up scores, differences in colonoscopy assessments can still affect patient diagnosis. Therefore, deep learning (DL) based automated UC grading methods could help reduce the operator subjectivity observed in complex UC scoring and improve diagnostic quality. Even though some DL methods have been devised, these methods 1) have below 80% classification accuracy and 2) must be rigorously assessed on multi-centre data. One major bottleneck of the technical development in this area is the lack of high-quality multi-centre datasets, which we aim to accomplish with this challenge. Furthermore, assessing the rigour and limitations of current and new methods will be the goal of this challenge.
Final dates follow the official ISBI 2026 schedule. Please refer to the ISBI Challenge page for any updates.
All submissions will be completed through the official submission portal available on Synapse
ISBI 2026 will be held at the ExCeL London Capital Suite