Dynamic Incentive Mechanism Design for COVID-19 Social Distancing

As countries enter the endemic phase of COVID-19, people's risk of exposure to the virus is greater than ever. There is a need to make more informed decisions in our daily lives on avoiding crowded places. Crowd monitoring systems typically require costly infrastructure. We propose a crowdsourced crowd monitoring platform which leverages user inputs to generate crowd counts and forecast location crowdedness. A key challenge for crowd-sourcing is a lack of incentive for users to contribute. We propose a Reinforcement Learning based dynamic incentive mechanism to optimally allocate rewards to encourage user participation.

Predictive Analytics for COVID-19 Social Distancing

The COVID-19 pandemic has disrupted the lives of millions across the globe. In Singapore, promoting safe distancing by managing crowds in public areas have been the cornerstone of containing the community spread of the virus. One of the most important solutions to maintain social distancing is to monitor the crowdedness of indoor and outdoor points of interest. Using Nanyang Technological University (NTU) as a testbed, we develop and deploy a platform that provides live and predicted crowd counts for key locations on campus to help users plan their trips in an informed manner, so as to mitigate the risk of community transmission. © 2021 International Joint Conferences on Artificial Intelligence. All rights reserved.

First nosocomial cluster of COVID-19 due to the Delta variant in a major acute care hospital in Singapore: investigations and outbreak response.

OBJECTIVES: The first large nosocomial cluster of coronavirus disease 2019 (COVID-19) in Singapore in April 2021 led to partial closure of a major acute care hospital. This study examined factors associated with infection among patients, staff and visitors; investigated the possible role of aerosol-based transmission; evaluated the effectiveness of BNT162.b2 and mRNA1273 vaccines; and described the successful containment of the cluster. METHODS: Close contacts of patients with COVID-19 and the affected ward were identified and underwent surveillance for severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection. Patient, staff and visitor cohorts were constructed and factors associated with infection were evaluated. Phylogenetic analysis of patient samples was performed. Ward air exhaust filters were tested for SARS-CoV-2. RESULTS: In total, there were 47 cases, comprising 29 patients, nine staff, six visitors and three household contacts. All infections were of the Delta variant. Ventilation studies showed turbulent air flow and swabs from air exhaust filters were positive for SARS-CoV-2. Vaccine breakthrough infections were seen in both patients and staff. Among patients, vaccination was associated with a 79% lower odds of infection with COVID-19 (adjusted odds ratio 0.21, 95% confidence interval 0.05-0.95). CONCLUSIONS: This cluster occurred despite enhancement of infection control measures that the hospital had undertaken at the onset of the COVID-19 pandemic. It was brought under control rapidly through case isolation, extensive contact tracing and quarantine measures, and led to enhanced use of hospital personal protective equipment, introduction of routine rostered testing of inpatients and staff, and changes in hospital infrastructure to improve ventilation within general wards.

High-fidelity simulation training with PPE may optimise resuscitation outcomes in the COVID-19 era

High-fidelity simulation training in CPR (CPR-HFST) could identify weaknesses in pre-COVID-19 Code Blue (CB) practices. Importantly, PPE donning may delay CPR thus worsening patient outcomes. We sought to determine the effect of our CPR-HFST on clinical practice. A retrospective review of CB events in a 1000-bedded hospital, pre- and intra-pandemic was conducted from 01/05/2019 to 30/10/2020. Onset of the pandemic was taken as 04/02/2020. CPR-HFST commenced in January 2020. The primary objective was to determine pre- and intra-pandemic response times. Intubation times, patient outcomes (quantified by CB survival rates and the CPC score), and incidence of HCW infection were our secondary objectives. The CCI score was used to stratify patients with similar comorbidities. Two-tailed Chi-square and Mann-Whitney tests were used for statistical comparisons, alpha = 0.05. 158 CB events were reviewed. Median response time was longer intra-pandemic compared to prepandemic;4.0 mins (IQR: 3-5) vs. 3.0 mins (IQR:1-4), p=0.0007. Cardiac rhythms were asystole (25.5%), PEA (53.8%), VT (5.7%), and VF (11.3%). 67.1% of patients required CPR, of which, 88.7% were intubated. There were no significant differences in median intubation times: 12.0 (prepandemic) vs. 11.0 mins (intra-pandemic), p=0.89. Survival to hospital discharge were similar;14.1% (pre-pandemic) vs. 21.4% (intra-pandemic), p=0.33. We did not find any significant differences in survival rates and CPC scores (Table 1). There were no HCW infections. Survival to hospital discharge rates of patients requiring in-hospital CPR may be lower intrapandemic;Miles et al reported 3.2% vs 12.8% respectively, p<0.01. These were significantly different compared to our intra-pandemic cohort (3.2% vs. 21.4%, p<0.01) but not in our pre-pandemic cohort (12.8% vs. 14.1%, p=0.82). Reasons for the differences are likely multifactorial. Nonetheless, in our experience and data, we believe CPR-HFST prevents deterioration in the standards of care and may help in optimising CPR outcomes. (Table Presented) .

AI-Empowered Decision Support for COVID-19 Social Distancing

The COVID-19 pandemic is one of the most severe challenges the world faces today. In order to contain the transmission of COVID-19, people around the world have been advised to practise social distancing. However, maintaining social distance is a challenging problem, as we often do not know beforehand how crowded the places we intend to visit are. In this paper, we demonstrate crowded.sg, an AI-empowered platform that leverages on Unmanned Aerial Vehicles (UAVs), crowdsourced images, and computer vision techniques to provide social distancing decision support.