A neural network algorithm for detection of GI angiectasia during small-bowel capsule endoscopy

• Published in: Gastrointestinal endoscopy Vol. 89; no. 1; pp. 189 - 194
• Main Authors: Leenhardt, Romain, Vasseur, Pauline, Li, Cynthia, Saurin, Jean Christophe, Rahmi, Gabriel, Cholet, Franck, Becq, Aymeric, Marteau, Philippe, Histace, Aymeric, Dray, Xavier, Sacher-Huvelin, Sylvie, Mesli, Farida, Leandri, Chloé, Nion-Larmurier, Isabelle, Lecleire, Stéphane, Gerard, Romain, Duburque, Clotilde, Vanbiervliet, Geoffroy, Amiot, Xavier, Philippe Le Mouel, Jean, Delvaux, Michel, Jacob, Pierre, Simon-Shane, Camille, Romain, Olivier
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.01.2019; Elsevier
• Subjects: Computer Science; Engineering Sciences; Human health and pathology; Hépatology and Gastroenterology; Image Processing; Life Sciences; Signal and Image processing; SB-CE; CNN; CAD-CAP; GIA; CAD; SB; Small Bowel; Angiodysplasia; Videocapsule
• ISSN: 0016-5107; 1097-6779; 1097-6779
• DOI: 10.1016/j.gie.2018.06.036

GI angiectasia (GIA) is the most common small-bowel (SB) vascular lesion, with an inherent risk of bleeding. SB capsule endoscopy (SB-CE) is the currently accepted diagnostic procedure. The aim of this study was to develop a computer-assisted diagnosis tool for the detection of GIA. Deidentified SB-CE still frames featuring annotated typical GIA and normal control still frames were selected from a database. A semantic segmentation images approach associated with a convolutional neural network (CNN) was used for deep-feature extractions and classification. Two datasets of still frames were created and used for machine learning and for algorithm testing. The GIA detection algorithm yielded a sensitivity of 100%, a specificity of 96%, a positive predictive value of 96%, and a negative predictive value of 100%. Reproducibility was optimal. The reading process for an entire SB-CE video would take 39 minutes. The developed CNN-based algorithm had high diagnostic performances, allowing detection of GIA in SB-CE still frames. This study paves the way for future automated CNN-based SB-CE reading softwares.