Impact of Residing in Below Median Household Income Districts on Outcomes in Patients with Advanced Barrett's Esophagus.

Background: Barrett's esophagus (BE) is a premalignant condition to esophageal adenocarcinoma (EAC). Low socioeconomic (SES) status adversely impacts care and outcomes in patients with EAC, but this has not been evaluated in BE. As the treatment of BE is similarly intensive, we aimed to evaluate the effect of SES on achieving complete eradication of intestinal metaplasia (CE-IM), dysplasia (CE-D) and development of invasive EAC. Methods: Our study was a retrospective cohort study. Consecutive patients between January 1, 2010, to December 31, 2018, referred for BE-associated high-grade dysplasia or intramucosal adenocarcinoma were included. Pre, intra and post-procedural data were collected. Household income data was collected from the 2016 census based on postal code region. Patients were divided into income groups relative to the 2016 median household income in Ontario. Multivariate regression was performed for outcomes of interest. Results: Four hundred and fifty-nine patients were included. Rate of CE-IM was similar between income groups. Fifty-five per cent (n = 144/264) versus 65% (n = 48/264) in the below and above-income groups achieved CE-D, respectively, P = 0.02. Eighteen per cent (n = 48/264) versus 11% (n = 22/195) were found to have invasive EAC during their treatment course in below and above-income groups, respectively, P = 0.04. Residing in a below-median-income district was associated with developing invasive EAC (Odds Ratio, [OR] 1.84, 95% confidence interval [CI] 1.01 to 3.35) and failure to achieve CE-D (OR 0.64, 95% CI 0.42 to 0.97). Conclusions: Residing in low-income districts is associated with worse outcomes in patients with advanced BE. Further research is needed to guide future initiatives to address the potential impact of SES barriers in the optimal care of BE.

Automated Detection of Bowel Preparation Scoring and Adequacy With Deep Convolutional Neural Networks.

Introduction: Adequate bowel preparation is integral to effective colonoscopy. Inadequate bowel preparation has been associated with reduced adenoma detection rate and increased post-colonoscopy colorectal cancer (PCCRC). As a result, the USMSTF recommends early interval reevaluation for colonoscopies with inadequate bowel preparation. However, bowel preparation documentation is highly variable with subjective interpretation. In this study, we developed deep convolutional neural networks (DCNN) to objectively ascertain bowel preparation. Methods: Bowel preparation scores were assigned using the Boston Bowel Preparation Scale (BBPS). Bowel preparation adequacy and inadequacy were defined as BBPS ≥2 and BBPS <2, respectively. A total of 38523 images were extracted from 28 colonoscopy videos and split into 26966 images for training, 7704 for validation, and 3853 for testing. Two DCNNs were created using a Densenet-169 backbone in PyTorch library evaluating BBPS score and bowel preparation adequacy. We used Adam optimiser with an initial learning rate of 3 × 10-4 and a scheduler to decay the learning rate of each parameter group by 0.1 every 7 epochs along with focal loss as our criterion for both classifiers. Results: The overall accuracy for BBPS subclassification and determination of adequacy was 91% and 98%, respectively. The accuracy for BBPS 0, BBPS 1, BBPS 2, and BBPS 3 was 84%, 91%, 85%, and 96%, respectively. Conclusion: We developed DCCNs capable of assessing bowel preparation adequacy and scoring with a high degree of accuracy. However, this algorithm will require further research to assess its efficacy in real-time colonoscopy.