Contributors: CHU Saint-Antoine AP-HP; Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU); Equipes Traitement de l'Information et Systèmes (ETIS - UMR 8051); Ecole Nationale Supérieure de l'Electronique et de ses Applications (ENSEA)-Centre National de la Recherche Scientifique (CNRS)-CY Cergy Paris Université (CY); Hôpital Avicenne AP-HP; Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP); Groupe Hospitalier Diaconesses Croix Saint-Simon; Hôpital Ambroise Paré AP-HP; Université de Versailles Saint-Quentin-en-Yvelines - UFR Sciences de la santé Simone Veil (UVSQ Santé); Université de Versailles Saint-Quentin-en-Yvelines (UVSQ); Hospices Civils de Lyon (HCL); ERGANEO (iPolyp Project), CY Initiative (SmartVideoColonoscopy project)Xavier Dray, Aymeric Histace and Romain Leenhardt are co-founders and shareholders of Augmented Endoscopy.Xavier Dray has received lecture fees from Bouchara Recordati, Fujifilm, Medtronic, MSD and Pfizer, and has acted as a consultant for Alfasigma, Boston Scientific, Norgine, and Pentax.Richard Makins has received travel and accommodation expenses from Intromedic to teach capsule endoscopy internationally and in the United Kingdom.Mark McAlindon has received research materials from Given Imaging, Jinshan Science, Intromedic, Ankon and has acted as a consultant for Medtronic.Robert Benamouzig is working for Alfasigma, Medtronics, and Bayer (board)
نبذة مختصرة : International audience ; Background and aimsCurrent artificial intelligence (AI)-based solutions for capsule endoscopy (CE) interpretation are proprietary. We aimed to evaluate an AI solution trained on a specific CE system (Pillcam®, Medtronic) for the detection of angiectasias on images captured by a different proprietary system (MiroCam®, Intromedic).Material and methodsAn advanced AI solution (Axaro®, Augmented Endoscopy), previously trained on Pillcam® small bowell images, was evaluated on independent datasets with more than 1200 Pillcam® and MiroCam® still frames (equally distributed, with or without angiectasias). Images were reviewed by experts before and after AI interpretation.ResultsSensitivity for the diagnosis of angiectasia was 97.4% with Pillcam® images and 96.1% with Mirocam® images, with specificity of 98.8% and 97.8%, respectively. Performances regarding the delineation of regions of interest and the characterization of angiectasias were similar in both groups (all above 95%). Processing time was significantly shorter with Mirocam® (20.7 ms) than with Pillcam® images (24.6 ms, p<0.0001), possibly related to technical differences between systems.ConclusionThis proof-of-concept study on still images paves the way for the development of resource-sparing, “universal” CE databases and AI solutions for CE interpretation.
_1.png)
Processing Request
No Comments.