¶

Part of the Endoscopic Vision Challenge¶

About¶

Early cancerous lesions in Barrett's Esophages (BE) are very hard to detect and are often overlooked during endoscopic surveillance of the esophagus. When the disease is detected early, the cancerous tissue can be surgically removed by endoscopic mucosal resection. This treatment knows very high curation rates, showing 93% of the patients in complete remission after 10 years. However, when a lesion is missed, and the disease is detected in a late stage,  the prognosis is grave with a five-year survival rate of approximately 15%. Therefore it is crucial that these cancerous lesions are identified at an early stage, where a Computer-Aided-Detection algorithm, that supports the gastroenterologist during endoscopic surveillance, is highly desirable.

In this sub-challenge, we focus on such supportive algorithms for the detection of early cancerous lesions in Barrets Esophagus. A set of 100 HD endoscopic images, annotated by five expert endoscopists, is employed for validation.

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:

Detection: Is the algorithm able to detect early cancerous lesions? 

*Annotation: *How well do the system delineations match the annotations of the medical experts?