Artificial intelligence (AI) has become a part of many image-based specialties, such
as radiology and pathology, as well as medical specialties in which “oscopy” is the
key to current practice. In gastroenterology, for example, AI is being explored as
an aid to endoscopists to visually distinguish precancerous lesions in upper and lower
endoscopy. Although there have been a variety of approaches to the employment of AI
for this purpose, deep-learning algorithms, which combine the extraction and classification
of image features using deep neural networks,1 have the capability of self-learning.