Optical Diagnosis of Neoplasia Using Artificial Intelligence

Computer‐Aided Diagnosis and cognitive forcing Computer‐aided diagnosis (CADx) forcolonoscopy is expected to improve physicians' ability to predict colorectal polyppathologies (optical diagnosis). However, recent randomized trials indicate thatcollaboration between humans and CADx yields lower performance than CADx alone.This suggests suboptimal human‐AI interaction due to users' under‐reliance on CADxadvice, favoring their inherent biases over critically assessing AI suggestions.Psychological interventions, such as cognitive forcing, aim to address this byencouraging crucial assessment of AI suggestions. One example, delayed display of CADxsuggestions may promote proactive thinking by giving endoscopists enough time to considerpolyp pathologies before receiving CADx suggestions, potentially leading to critical andoptimal human‐AI interaction. The effectiveness of such cognitive forcing was observed inexperimental studies of AI for mammography reading.However, despite its potential, no studies have evaluated the impact of suchinterventions on optical diagnosis accuracy in colonoscopy. To investigate the value ofpsychological intervention in optical diagnosis, the investigators will conduct anex‐vivo randomised controlled study.Study aim:This is an ex‐vivo randomised controlled study. The hypothesis of our study is thatcognitive forcing by delaying display of CADx suggestions facilitates physicians'critical thinking, leading to better clinical outcomes in optical diagnosis incolonoscopy.Comercially CADx device and video details:In this study, the investigators are going to use the comercially available CADx system(GI Genius CADx, manufactured by Cosmo Intelligent Medical Devices and distributed byMedtronic Corp). The GI Genius CADx highlights the suspicious area for polyps on‐screenwith bounding boxes and provides optical diagnosis prediction (i.e. adenoma,non‐adenoma). All the endoscopists will be given the information that the CADx tool usedin the study is GI Genius with a link to their product overview.The investigators will collect 100 colonoscopy videos of 100 different diminutive polypsfrom the Polyp Image BAnk database (PIBAdb). In accordance with the real‐worldprevalence, 65 polyps will be neoplastic while the remaining 35 will be non‐neoplastic.The duration of each video will be adjusted to contain 15 second appearance of the lesionincluding wite light (WL) and narrow band imaging (NBI). All the polyps should be < = 5mm. The original videos were recorded without having any CADx interaction.PIBAdb contains 507 videos of diminutive colorectal polyps with WL and NBI, of which 231have a duration of 15 seconds or more. All the videos contain polyps with theirhistopathology available (i.e. adenoma, sessile serrated lesions, traditional serratedadenomas, invasive, hyperplastic and non‐neoplastic). The polyps that have no histologycategory are going to be excluded from the study.The investigators will use this database as a pool to select the 100 study videos. First,the investigators are going to split the pool of data into two groups according to polyphistology: one pool containing neoplasia (i.e. adenoma, sessile serrated lesions andtraditional serrated adenoma) and the other containing non‐neoplasia (i.e. hyperplasticand non‐epithelial neoplastic). In each pool, 20 videos will be randomly selected as afirst step. These 20 videos will be assessed if they meet our inclusion and exclusioncriteria (see below), leaving only eligible videos. This selection process will berepeatedly done until the investigators collect 65 videos of polyps with neoplasia and 35videos of polyps with non‐neoplasia. The figure below shows how the video selectionprocess takes place.GI Genius´s specification The GI Genius system processes the videos at 50‐60 frames persecond and in high‐definition format to have a good quality of the videos. To process thevideos, the GI Genius splits them into two different streams by dedicated video card. Onestream is transmitted to the first path to the output without any processing of the AIalgorithm. The second stream is sent to the AI algorithm (i.e. second path) and afterappropriate computation, if there is a polyp present on‐screen appears an overlay in theoutput highlighting the polyp. The system was designed to work on unaltered WL videostreams, but it works under both WL and NBI similarly.The transmission of the videostream to GI Genius is through a Serial Digital Interface (SDI) cable, acquiring theoutput stream from the video displaying in a computer. Finally, the SDI output stream istransmitted to the monitor containing the original video stream with additional markerssuperimposed on it.How to make CADx overlaid videos In the present study, the selected videos of colorectallesions will be stored in our high‐spec computer system first. Then, these videos will betransmitted to GI Genius with an SDI cable. A capture card (DeckLink 8K Pro Mini,Blackmagic) is integrated into the computer system that converts the recorded videooutput to SDI signal, which allows GI Genius to process the transmitted videos.All 100 selected videos will be processed by GI Genius in two different ways. In thefirst set, only the frames from the last 3 seconds of each 15‐second video will beprocessed, and CADx suggestions will appear only in that final 3‐second segment. Thisfirst set will be shown to endoscopists who are allocated to the intervention arm. In thesecond set, all frames from the entire 15‐second duration will be processed, and CADxsuggestions will be overlaid on every frame. This first set will be shown to endoscopistswho are allocated to the control arm.