An LSTM1 Model for Managing Epidemics: Using Population Mobility and Vulnerability for Forecasting COVID-19 Hospital Admissions

Worldwide epidemics, such as corona virus disease 2019 (COVID-19), cause unprecedented challenges for society and its healthcare systems. Governments attempt to mitigate those challenges by either reducing healthcare demand ("flattening the curve" by imposing restrictions, e.g., on travel or social gatherings) or by increasing healthcare capacity, for example, by canceling elective procedures or setting up field hospitals. To implement these mitigation procedures efficiently, accurate and timely forecasts of the epidemic's progression are necessary. In this paper, we develop an innovative forecasting methodology based on the ideas of long short-term memory (LSTM) recurrent neural networks. LSTM models are shown to outperform traditional forecasting models, especially when the relationship between input and output is complex and not available in closed form. However, whereas LSTM models perform well for data that changes dynamically over time, one shortcoming is that they are not directly applicable when the data also includes static, nontemporal components. In this work, we propose an LSTM+ model that overcomes this limitation. Our model leverages a private partnership with a mobile data company in order to capture population mobility (using mobility indices derived from mobile device data), which allows us to anticipate an epidemic's spread early and accurately. In addition, we also leverage a public partnership with a consortium of hospitals. Using hospital admissions (rather than, say, positive caseload) results in an unbiased measure of the severity of an epidemic because patients seek and are admitted to hospital care only when symptoms worsen beyond a critical point. We illustrate the effectiveness of our method on forecasting COVID-19 for a major U.S. metropolitan area where it has aided decision makers of the emergency policy group. Our model improves the predictive accuracy of hospital admission by a factor of 2.5x as compared with competing models in the same analytical space.

Impact of mobility based network on COVID-19 spread

COVID-19 has had a strong impact on this world. With the spreading of the virus and the implementation of various mitigation measures, the pandemic has indubitably upended our way of living. Research indicates that mobility is one of the key reasons of the spread. The purpose of this paper is to provide a suitable mobility measure based on intra-county and inter-county movements on the spreading of COVID-19 in the United States. Deviating from the extant research, which measures mobility by the average distance people travel, we operationalise mobility by the number of trips made. We further weigh them based on the current caseload, as the spread will not only depend on how many people are moving but also the proportion of infectious people within them. We also distinguish such trips based on their origin and destination, as that may help in taking appropriate policy decisions for intervention.

Pandemic-proof teaching: Blended learning infrastructure to support a pivot to hybrid/online pedagogy

The use of a blended learning model in an undergraduate anatomy & physiology course is described, along with the impact of this pedagogy during the transition to remote teaching due to COVID-19. The case study involves a multi-year implementation aimed at improving student preparedness and engagement through a variety of online tools, including adaptive learning activities, interactive homework, and formative assessments. Data gathering involved courseware analytics on student performance, student surveys, and course evaluations. The use of blended learning in the laboratory represents an innovative reform to traditional instruction. The model has yielded a progressive and sustained improvement in learning, accompanied by an overall decrease in attrition. Results also suggest that the blended pedagogical foundation provided the flexibility to successfully transition to remote/hybrid teaching and learning during the pandemic. © 2022 Elsevier Ltd. All rights reserved.

Effect of COVID-19 lockdown on sleep behavior and screen exposure time: an observational study among Indian school children

Lockdown is an important measure that has been globally adopted to reduce the spread of the contagious disease caused by SARS CoV-2. The imposed schedule and confinement led to extensive use of digital media and rise in sedentary activity drastically. The escalated duration of screen exposure causes disruption in sleep behavior. An online survey was conducted to comprehend the effect of lockdown on sleep behavior and screen exposure time on school children. Screen exposure time involved with various electronic gadgets was also analyzed. It was observed that the social jet lag and sleep debt were significantly less during lockdown than before it. Inertia during lockdown significantly increased. The difference between screen exposure time on weekdays before lockdown and weekends during lockdown was identified to be the highest. Three clusters based on sleep behavior and duration of screen time were identified of which Cluster 2 revealed simultaneous existence of high sleep duration and screen time. These baseline data on sleep parameters and duration of exposure to the screen will help us in devising approaches to mitigate the evident disruption this unprecedented phase has brought about.