Objectives: This scoping review was conducted to determine the barriers and enablers associated with the acceptance of artificial intelligence/machine learning (AI/ML)-enabled innovations into radiology practice from a physician’s perspective. Methods: A systematic search was performed using Ovid Medline and Embase. Keywords were used to generate refined queries with the inclusion of computer-aided diagnosis, artificial intelligence, and barriers and enablers. Three reviewers assessed the articles, with a fourth reviewer used for disagreements. The risk of bias was mitigated by including both quantitative and qualitative studies. Results: An electronic search from January 2000 to 2023 identified 513 studies. Twelve articles were found to fulfill the inclusion criteria: qualitative studies (n = 4), survey studies (n = 7), and randomized controlled trials (RCT) (n = 1). Among the most common barriers to AI implementation into radiology practice were radiologists’ lack of acceptance and trust in AI innovations; a lack of awareness, knowledge, and familiarity with the technology; and perceived threat to the professional autonomy of radiologists. The most important identified AI implementation enablers were high expectations of AI’s potential added value; the potential to decrease errors in diagnosis; the potential to increase efficiency when reaching a diagnosis; and the potential to improve the quality of patient care. Conclusions: This scoping review found that few studies have been designed specifically to identify barriers and enablers to the acceptance of AI in radiology practice. The majority of studies have assessed the perception of AI replacing radiologists, rather than other barriers or enablers in the adoption of AI. To comprehensively evaluate the potential advantages and disadvantages of integrating AI innovations into radiology practice, gathering more robust research evidence on stakeholder perspectives and attitudes is essential.

Polypectomy is an essential endoscopic skill. Training in polypectomy has been identified as a major deficiency for endoscopists worldwide as polypectomy occurs ad hoc during a colonoscopy when a polyp is detected, and a lack of standardized curricula. Augmented reality (AR), which superimposes computer-generated images on a user’s view of the world, can address these gaps by standardizing encounters with polyps while completing simulated procedures and enabling polypectomy-specific teaching. Evaluate the impact of a simulation-based augmented reality curriculum on polypectomy performance among novice endoscopists. This study includes two cohorts of participants from 2019 to 2020. In 2019, participants were randomized into either: (1) a control curriculum, involving 6 hours of simulation-based training (SBT) supplemented by expert feedback, interlaced with 4 hours of small group teaching on the theory of colonoscopy; or (2) the augmented reality curriculum (ARC), in involving the same curriculum with integrated AR, wherein participants engaged with an AR-enhanced video demonstrating relevant therapeutic and pathologic details during polypectomy. The SBT for all participants involved a progressive curriculum starting on a bench-top model and then moving to the EndoVR® virtual reality simulator. The primary outcome was polypectomy-specific performance using the Direct Observation of Polypectomy Skills (DOPyS) tool during a simulated polypectomy after training, with a maximum score of 100. Demographic characteristics are summarized in Table. In 2019, 21 novice endoscopists were enrolled. Immediately after training, the mean DOPyS score among ARC group participants was 76.2 (SD=17.9) compared to 71.8 (SD=13.2) among control group participants (Figure). In this interim analysis, there was no significant difference between groups. Data analysis will be completed after 2020 participants complete the study. Interim results show a trend towards improved polypectomy performance with no significant difference. The results of this study have the potential to impact polypectomy education among novices. Simulation-based AR interventions may allow learners to progress towards achieving competency in polypectomy in a risk-free environment prior to first patient contact. Demographic characteristics of 2019 participants Demographic characteristics of 2019 participants Polypectomy performance before and after training. Polypectomy performance before and after training. None

Background: Magnification endoscopy and magnification narrow-band imaging are image enhanced endoscopy technologies that may allow for the diagnosis of advanced neoplasia in the GI tract on the basis of imaging characteristics. Recently, the Unified Magnifying Endoscopic Classification (UMEC) has been developed, which unified the criteria for the esophagus, stomach, and colon. UMEC divides optical diagnosis into one of the three categories: non-neoplastic, intramucosal neoplasia, and deep submucosal invasive cancer. Purpose: The objective of this study is to educate North American endoscopists on the use of the UMEC schema, and to ascertain performance of the UMEC framework among North American endoscopists. Method: Using UMEC, five North American endoscopists (>1000 procedures) without prior training in magnifying endoscopy independently diagnosed previously collected endoscopic image set of the esophagus, stomach, and colon. The endoscopists were trained on the use of UMEC via an eleven-minute training video with exemplars of each element of UMEC from esophagus, stomach, and colon. All endoscopists were blinded to white-light and non-magnifying NBI findings as well as histopathological diagnosis. The diagnostic performance of UMEC was assessed while using the gold standard histopathology as a reference. Result(s): A total of 299 gastrointestinal lesions (77 esophagus, 92 stomach, and 130 colon) were assessed using UMEC. For esophageal squamous cell carcinoma, the sensitivity, specificity, and accuracy for all 5 endoscopists ranged from 65.2% (95% CI: 50.9–77.9) to 87.0% (95% CI: 75.3–94.6), 77.4% (95% CI: 60.9–89.6) to 96.8% (95% CI: 86.8–99.8), and 75.3% to 87.0%, respectively. For gastric adenocarcinoma, the sensitivity, specificity, and accuracy for all 5 endoscopists ranged from 94.9% (95% CI: 85.0–99.1) to 100%, 52.9% (95% CI: 39.4–66.2) to 92.2% (95% CI: 82.7–97.5), and 73.3% to 93.3%, respectively. For colorectal adenocarcinoma, the sensitivity, specificity, and accuracy for all 5 endoscopists ranged from 76.2% (95% CI: 62.0–87.3) to 83.3% (95% CI: 70.3–92.5), 89.7% (95% CI: 82.1–94.9) to 97.7% (95% CI: 93.1–99.6), and 86.8% to 90.7%, respectively. Image: Conclusion(s): UMEC is a simple and practical classification that can be used to introduce and educate endoscopists to magnification narrow-band imaging and optical diagnosis. Please acknowledge all funding agencies by checking the applicable boxes below: CAG Disclosure of Interest: M. R. A. Fujiyoshi Grant / Research support from: 2022 CAG/AbbVie Education Research Grant, Y. Fujiyoshi: None Declared, N. Gimpaya: None Declared, R. Bechara: None Declared, T. Jeyalingam: None Declared, N. Calo: None Declared, N. Forbes: None Declared, R. Khan: None Declared, M. Atalla: None Declared, A. Toshimori: None Declared, Y. Shimamura: None Declared, M. Tanabe: None Declared, J. Mosko: None Declared, H. Inoue: None Declared, S. Grover: None Declared