Analysis of mobility based COVID-19 epidemic model using Federated Multitask Learning
Open Access
- 1 January 2022
- journal article
- research article
- Published by American Institute of Mathematical Sciences (AIMS) in Mathematical Biosciences and Engineering
- Vol. 19 (10), 9983-10005
- https://doi.org/10.3934/mbe.2022466
Abstract
Aggregating a massive amount of disease-related data from heterogeneous devices, a distributed learning framework called Federated Learning(FL) is employed. But, FL suffers in distributing the global model, due to the heterogeneity of local data distributions. To overcome this issue, personalized models can be learned by using Federated multitask learning(FMTL). Due to the heterogeneous data from distributed environment, we propose a personalized model learned by federated multitask learning (FMTL) to predict the updated infection rate of COVID-19 in the USA using a mobility-based SEIR model. Furthermore, using a mobility-based SEIR model with an additional constraint we can analyze the availability of beds. We have used the real-time mobility data sets in various states of the USA during the years 2020 and 2021. We have chosen five states for the study and we observe that there exists a correlation among the number of COVID-19 infected cases even though the rate of spread in each case is different. We have considered each US state as a node in the federated learning environment and a linear regression model is built at each node. Our experimental results show that the root-mean-square percentage error for the actual and prediction of COVID-19 cases is low for Colorado state and high for Minnesota state. Using a mobility-based SEIR simulation model, we conclude that it will take at least 400 days to reach extinction when there is no proper vaccination or social distance.Keywords
This publication has 37 references indexed in Scilit:
- The socio-economic implications of the coronavirus pandemic (COVID-19): A reviewInternational Journal of Surgery, 2020
- WHO Declares COVID-19 a Pandemic2020
- Numerical modeling and theoretical analysis of a nonlinear advection-reaction epidemic systemComputer Methods and Programs in Biomedicine, 2020
- Analysis and Nonstandard Numerical Design of a Discrete Three-Dimensional Hepatitis B Epidemic ModelMathematics, 2019
- Reproduction numbers of infectious disease modelsInfectious Disease Modelling, 2017
- Market-based control of epidemicsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2011
- Convex multi-task feature learningMachine Learning, 2008
- Optimal and sub-optimal quarantine and isolation control in SARS epidemicsMathematical and Computer Modelling, 2007
- Convex Multi-Task Feature LearningSSRN Electronic Journal, 2007
- A contribution to the mathematical theory of epidemicsProceedings of the Royal Society of London. Series A, Containing Papers of a Mathematical and Physical Character, 1927