Development of a Natural Language Processing Pipeline for Calculating Colonoscopy Quality Indicators: Comparison of Manual Review and Natural Language Processing (Preprint)

BACKGROUND Manual data extraction for colonoscopy quality indicators is time- and labor-intensive. Natural language processing (NLP), a computer-based linguistics and technique, offers the automation of reporting from unstructured free text reports to extract important clinical information. The application of information extraction using NLP includes identification of clinical information such as adverse events and clinical work optimization such as quality control and patient management. OBJECTIVE We developed a natural language processing pipeline to manage Korean–English colonoscopy reports and evaluated its performance on automatically assessing adenoma detection rate (ADR), sessile serrated lesion detection rate (SDR), and surveillance interval (SI). METHODS The NLP tool was developed using 2000 screening colonoscopy records (1425 pathology reports) at Seoul National University Hospital Gangnam Center. Tests were performed on another 1,000 colonoscopy records to compare a manual review (MR) by five human annotators and the NLP pipeline. Additionally, data from 54,562 colonoscopies of 12,264 patients (aged ≥50 years) from 2010 to 2019 were analyzed using the NLP pipeline for colonoscopy quality indicators. RESULTS The overall accuracy of the test dataset was 95.8% (958/1000) for NLP vs. 93.1% (931/1000) for MR (P=.008). The mean total ADR in the test set was 46.8% (468/1000) with NLP vs. 47.2% (472/1000) with MR. The mean total SDR was 6.4% (64/1000) with NLP vs. 6.5% (65/1000) with MR. Calculating the SI revealed a similar performance between both methods. The mean ADR and SDR of the 25 endoscopists in the 10-year dataset were 42.0% (881/2098) and 3.3% (69/2098), respectively, indicating wide individual variability (16.3% (263/1615)–56.2% (1014/1936) in ADR and 0.4% (6/1615)–6.6% (124/1876) in SDR). The SI recommendation suggested a large difference in ADR and SDR based on the endoscopist’s performance. CONCLUSIONS The NLP pipeline can accurately and automatically calculate ADR, SDR, and SI from a multi-language colonoscopy report. It could be an important tool for improving colonoscopy quality and clinical decision support. CLINICALTRIAL This study was approved by the Institutional Review Board of SNUH (IRB 1909-093-670).

The Impact of Face Shields on the Quality of Colonoscopy During the COVID-19 Pandemic

Background: The coronavirus disease 2019 (COVID-19) has become a global pandemic. Healthcare workers are at a higher risk for exposure to COVID-19 infection than the general population. During the COVID-19 pandemic, endoscopists are recommended to wear personal protective equipment (PPE), including face shields, to prevent COVID-19 transmission. However, to the best of our knowledge, there are no reports on the impact of face shields on the quality of colonoscopy. This study aimed to determine whether the use of PPE, including face shields, affects the quality of colonoscopy during the COVID-19 pandemic.Methods: We retrospectively reviewed the medical records of patients who underwent screening or surveillance colonoscopy performed at Dong-A University Hospital between June 2020 and March 2021 during the COVID-19 pandemic. Endoscopists wore isolation gowns, disposable gloves, and KF94 masks from June 2020 to October 2020. From November 2020, endoscopists additionally wore face shields. Therefore, we compared the colonoscopy quality indicators during the 5 months without the use of face shields and the 5 months with the use of face shields. We calculated the overall adenoma detection rates (ADRs) of the group using face shields and the group not using face shields. Further, the polyp detection rate (PDR), sessile serrated lesion detection rate (SSLDR), advanced neoplasia detection rate (ANDR), polyp per colonoscopy, and adenoma per colonoscopy were calculated for each group.Results: In total, 1,359 patients were included in the study; the face shield and non-face shield groups comprised 679 and 680 patients, respectively. We found no statistically significant differences in the PDR (49.04 vs. 52.50%, p=0.202), ADR (38.59 vs. 38.97%, p=0.884) SSPDR (1.91 vs. 1.32%, p=0.388), and ANDR (3.98 vs. 3.97%, p=0.991) between the groups. In both the experienced endoscopist group and trainee endoscopist group, there was no difference in the colonoscopy quality indicators between the groups of patients examined by endoscopists with and without face shields.Conclusions: The quality indicators of colonoscopy were not affected by face shields during the COVID-19 pandemic.

The analysis of multilevel factors affecting adenoma detection rates for colonoscopies: a large-scale retrospective study

Background Adenoma detection rate (ADR) is a validated primary quality indicator for colonoscopy procedures. However, there is growing concern over the variability associated with ADR indicators. Currently, the factors that influence ADRs are not well understood. Aims In this large-scale retrospective study, the impact of multilevel factors on the quality of ADR-based colonoscopy was assessed. Methods A total of 10,788 patients, who underwent colonoscopies performed by 21 endoscopists between January 2019 and December 2019, were retrospectively enrolled in this study. Multilevel factors, including patient-, procedure-, and endoscopist-level characteristics were analyzed to determine their relationship with ADR. Results The overall ADR was 20.21% and ranged from 11.4 to 32.8%. Multivariate regression analysis revealed that higher ADRs were strongly correlated with the following multilevel factors: patient age per stage (OR 1.645; 95% CI 1.577–1.717), male gender (OR 1.959; 95% CI 1.772–2.166), sedation (OR 1.402; 95% CI 1.246–1.578), single examiner colonoscopy (OR 1.330; 95% CI 1.194–1.482) and senior level endoscopists (OR 1.609; 95% CI 1.449–1.787). Conclusion The ADR is positively correlated with senior level endoscopists and single examiner colonoscopies in patients under sedation. As such, these procedure- and endoscopist-level characteristics are important considerations to improve the colonoscopy quality.