Artificial Intelligence and Device-Assisted Enteroscopy: Automatic Detection of Enteric Protruding Lesions Using a Convolutional Neural Network

The endoscopic evaluation of the small bowel has historically been technically difficult. Capsule endoscopy has revolutionized the approach to patients with suspected small bowel diseases. Nevertheless, this non-invasive technique has limitations, particularly its inability to provide tissue sampling. This is a problem particularly when in the presence of a suspected neoplastic lesion. Small bowel tumors represent almost 10% of all gastrointestinal tumors, and device-assisted enteroscopy plays an important role in the work up of these patients. 

Our group has developed an AI algorithm for the automatic detection of protruding lesions within the small bowel during device-assisted enteroscopy. This aims to complement our panel of AI algorithms for the application to several endoscopic modalities.

The link to the article can be found here.

In Press // Clinical and Translational Gastroenterology 2022 Aug 1;13(8):e00514 // August 2022