Bracing for impact: operational upshots from the National Centre for Infectious Diseases Screening Centre (Singapore) during the COVID-19 outbreak

Diagnostic Performance of a Deep Learning Model Deployed at a National COVID-19 Screening Facility for Detection of Pneumonia on Frontal Chest Radiographs

(1) Background: Chest radiographs are the mainstay of initial radiological investigation in this COVID-19 pandemic. A reliable and readily deployable artificial intelligence (AI) algorithm that detects pneumonia in COVID-19 suspects can be useful for screening or triage in a hospital setting. This study has a few objectives: first, to develop a model that accurately detects pneumonia in COVID-19 suspects; second, to assess its performance in a real-world clinical setting; and third, by integrating the model with the daily clinical workflow, to measure its impact on report turn-around time. (2) Methods: The model was developed from the NIH Chest-14 open-source dataset and fine-tuned using an internal dataset comprising more than 4000 CXRs acquired in our institution. Input from two senior radiologists provided the reference standard. The model was integrated into daily clinical workflow, prioritising abnormal CXRs for expedited reporting. Area under the receiver operating characteristic curve (AUC), F1 score, sensitivity, and specificity were calculated to characterise diagnostic performance. The average time taken by radiologists in reporting the CXRs was compared against the mean baseline time taken prior to implementation of the AI model. (3) Results: 9431 unique CXRs were included in the datasets, of which 1232 were ground truth-labelled positive for pneumonia. On the “live” dataset, the model achieved an AUC of 0.95 (95% confidence interval (CI): 0.92, 0.96) corresponding to a specificity of 97% (95% CI: 0.97, 0.98) and sensitivity of 79% (95% CI: 0.72, 0.84). No statistically significant degradation of diagnostic performance was encountered during clinical deployment, and report turn-around time was reduced by 22%. (4) Conclusion: In real-world clinical deployment, our model expedites reporting of pneumonia in COVID-19 suspects while preserving diagnostic performance without significant model drift.

One Year on: An Overview of Singapore's Response to COVID-19-What We Did, How We Fared, How We Can Move Forward

Background—One year has passed since the first COVID-19 case in Singapore. This scoping review commemorates Singaporean researchers that have expanded the knowledge on this novel virus. We aim to provide an overview of healthcare-related articles published in peer-reviewed journals, authored by the Singapore research community about COVID-19 during the first year of the pandemic. Methods—This was reported using the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) protocol. It included healthcare-related articles about COVID-19 published between 23 January 2020 and 22 January 2021 with a Singapore-affiliated author. MEDLINE, Embase, Scopus, Web of Science, CINAHL, PsycINFO, Google Scholar, and local journals were searched. The articles were screened independently by two reviewers. Results—The review included 504 articles. Most of the articles narrated the changes to hospital practice (210), while articles on COVID-19 pathology (94) formed most of the non-narrative papers. Publications on public health (61) and the indirect impacts to clinical outcomes (45) were other major themes explored by the research community. The remaining articles detailed the psychological impact of the pandemic (35), adaptations of medical education (30), and narratives of events (14). Conclusion—Amidst a resurgence of community cases involving variant COVID-19 strains, the resources from the research community will provide valuable guidance to navigate these uncertain times.

Dynamic emergency department response to the evolving COVID-19 pandemic: the experience of a tertiary hospital in Singapore

The coronavirus disease 2019 (COVID‐19) pandemic has placed large stressors on emergency departments (EDs) worldwide. As the pandemic progressed, EDs faced changing patient epidemiology and numbers. Our ED needed to rapidly transform to deal with the risk of COVID‐19. Having limited floor space, we opted for a phased, dynamic response that allowed us to adapt the ED multiple times as the epidemiology of the pandemic evolved. The principles behind our response include guiding ED operations with data, enhancing infection control practices, and being prepared to transform areas of the ED to care for different groups of patients. Our experience can serve to guide other EDs in planning their response to surge capacity and ED operations during such pandemics.

Bracing for impact: operational upshots from the National Centre for Infectious Diseases Screening Centre (Singapore) during the COVID-19 outbreak

To combat the ongoing COVID‐19 pandemic, Singapore has adopted a rigorous screening approach that involves aggressive contact tracing, rapid isolation of confirmed or suspect cases, and immediate ring‐fencing of emerging local clusters and hotspots. Our screening centre team has been involved in running Singapore's designated screening centre since the end of January this year. With a well‐defined blueprint and substantial pre‐outbreak preparatory work, initial operations at our screening centre commenced within a day on activation and full operational status was attained in 3 days. As of 8 April 2020, the screening centre had screened more than 14,000 patients. We have adopted a “whole‐of‐hospital” approach, enlisting the help from other departments and subspecialties to augment manpower. Meticulous infrastructure planning to facilitate patient flow and strict measures to prevent nosocomial transmission and occupational exposure were instituted to safeguard both the staff and patients. This paper aims to describe our key takeaways in the course of operations and discuss the challenges encountered.