Democratizing Flexible Endoscopy Training: Noninferiority Randomized Trial Comparing a Box-Trainer vs a Virtual Reality Simulator to Prepare for the Fundamental of Endoscopic Surgery Exam.

BACKGROUND: A considerable number of surgical residents fail the mandated endoscopy exam despite having completed the required clinical cases. Low-cost endoscopy box trainers (BTs) could democratize training; however, their effectiveness has never been compared with higher-cost virtual reality simulators (VRSs). STUDY DESIGN: In this randomized noninferiority trial, endoscopy novices trained either on the VRS used in the Fundamental of Endoscopic Surgery manual skills (FESms) exam or a validated BT-the Basic Endoscopic Skills Training (BEST) box. Trainees were tested at fixed timepoints on the FESms and on standardized ex vivo models. The primary endpoint was FESms improvement at 1 week. Secondary endpoints were FESms improvement at 2 weeks, FESms pass rates, ex vivo tests performance, and trainees' feedback. RESULTS: Seventy-seven trainees completed the study. VRS and BT trainees showed comparable FESms improvements (25.16 ± 14.29 vs 25.58 ± 11.75 FESms points, respectively; p = 0.89), FESms pass rates (76.32% vs 61.54%, respectively; p = 0.16) and total ex vivo tasks completion times (365.76 ± 237.56 vs 322.68 ± 186.04 seconds, respectively; p = 0.55) after 1 week. Performances were comparable also after 2 weeks of training, but FESms pass rates increased significantly only in the first week. Trainees were significantly more satisfied with the BT platform (3.97 ± 1.20 vs 4.81 ± 0.40 points on a 5-point Likert scale for the VRS and the BT, respectively; p < 0.001). CONCLUSIONS: Simulation-based training is an effective means to develop competency in endoscopy, especially at the beginning of the learning curve. Low-cost BTs like the BEST box compare well with high-tech VRSs and could help democratize endoscopy training.

Population-Based Prevalence of Gastrointestinal Abnormalities at Colon Capsule Endoscopy.

BACKGROUND & AIMS: The population prevalence of gastrointestinal (GI) disease is unclear and difficult to assess in an asymptomatic population. The aim of this study was to determine prevalence of GI lesions in a largely asymptomatic population undergoing colon capsule endoscopy (CCE). METHODS: Participants aged between 50-75 years were retrieved from the Rotterdam Study, a longitudinal epidemiological study, between 2017-2019. Participants received CCE with bowel preparation. Abnormalities defined as clinically relevant were Barrett segment >3cm, severe ulceration, polyp >10 mm or ≥3 polyps in small bowel (SB) or colon, and cancer. RESULTS: Of 2800 invited subjects, 462 (16.5%) participants (mean age 66.8 years, female 53.5%) ingested the colon capsule. A total of 451 videos were analyzed, and in 94.7% the capsule reached the descending colon. At least 1 abnormal finding was seen in 448 (99.3%) participants. The prevalence of abnormalities per GI segment, and the most common type of abnormality, were as follows: Esophageal 14.8% (Barrett's esophagus <3 cm in 8.3%), gastric 27.9% (fundic gland polyps in 18.1%), SB abnormalities 33.9% (erosions in 23.8%), colon 93.3% (diverticula in 81.2%). A total of 54 participants (12%) had clinically relevant abnormalities, 3 (0.7%) in esophagus/stomach (reflux esophagitis grade D, Mallory Weiss lesion and severe gastritis), 5 (1.1%) in SB (polyps > 10 mm; n = 4, severe ulcer n = 1,) and 46 (10.2%) in colon (polyp > 10 mm or ≥3 polyps n = 46, colorectal cancer n = 1). CONCLUSIONS: GI lesions are very common in a mostly asymptomatic Western population, and clinically relevant lesions were found in 12% at CCE. These findings provide a frame of reference for the prevalence rates of GI lesions in the general population.

Computer versus cardiologist: Is a machine learning algorithm able to outperform an expert in diagnosing a phospholamban p.Arg14del mutation on the electrocardiogram?

BACKGROUND: Phospholamban (PLN) p.Arg14del mutation carriers are known to develop dilated and/or arrhythmogenic cardiomyopathy, and typical electrocardiographic (ECG) features have been identified for diagnosis. Machine learning is a powerful tool used in ECG analysis and has shown to outperform cardiologists. OBJECTIVES: We aimed to develop machine learning and deep learning models to diagnose PLN p.Arg14del cardiomyopathy using ECGs and evaluate their accuracy compared to an expert cardiologist. METHODS: We included 155 adult PLN mutation carriers and 155 age- and sex-matched control subjects. Twenty-one PLN mutation carriers (13.4%) were classified as symptomatic (symptoms of heart failure or malignant ventricular arrhythmias). The data set was split into training and testing sets using 4-fold cross-validation. Multiple models were developed to discriminate between PLN mutation carriers and control subjects. For comparison, expert cardiologists classified the same data set. The best performing models were validated using an external PLN p.Arg14del mutation carrier data set from Murcia, Spain (n = 50). We applied occlusion maps to visualize the most contributing ECG regions. RESULTS: In terms of specificity, expert cardiologists (0.99) outperformed all models (range 0.53-0.81). In terms of accuracy and sensitivity, experts (0.28 and 0.64) were outperformed by all models (sensitivity range 0.65-0.81). T-wave morphology was most important for classification of PLN p.Arg14del carriers. External validation showed comparable results, with the best model outperforming experts. CONCLUSION: This study shows that machine learning can outperform experienced cardiologists in the diagnosis of PLN p.Arg14del cardiomyopathy and suggests that the shape of the T wave is of added importance to this diagnosis.