Deep learning via LSTM models for COVID-19 infection forecasting in India

被引:82
作者
Chandra, Rohitash [1 ]
Jain, Ayush [2 ]
Chauhan, Divyanshu Singh [3 ]
机构
[1] Univ New South Wales, Sch Math & Stat, Transit Artificial Intelligence Res Grp, Sydney, NSW, Australia
[2] Indian Inst Technol Guwahati, Dept Elect & Elect Engn, Gauhati, Assam, India
[3] Indian Inst Technol Guwahati, Dept Mech Engn, Gauhati, Assam, India
关键词
RECURRENT NEURAL-NETWORKS; FINITE-STATE AUTOMATA; HEALTH; PREDICTION;
D O I
10.1371/journal.pone.0262708
中图分类号
O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];
学科分类号
07 ; 0710 ; 09 ;
摘要
The COVID-19 pandemic continues to have major impact to health and medical infrastructure, economy, and agriculture. Prominent computational and mathematical models have been unreliable due to the complexity of the spread of infections. Moreover, lack of data collection and reporting makes modelling attempts difficult and unreliable. Hence, we need to re-look at the situation with reliable data sources and innovative forecasting models. Deep learning models such as recurrent neural networks are well suited for modelling spatiotemporal sequences. In this paper, we apply recurrent neural networks such as long short term memory (LSTM), bidirectional LSTM, and encoder-decoder LSTM models for multi-step (short-term) COVID-19 infection forecasting. We select Indian states with COVID-19 hotpots and capture the first (2020) and second (2021) wave of infections and provide two months ahead forecast. Our model predicts that the likelihood of another wave of infections in October and November 2021 is low; however, the authorities need to be vigilant given emerging variants of the virus. The accuracy of the predictions motivate the application of the method in other countries and regions. Nevertheless, the challenges in modelling remain due to the reliability of data and difficulties in capturing factors such as population density, logistics, and social aspects such as culture and lifestyle.
引用
收藏
页数:28
相关论文
共 99 条
[1]   Forecasting the spread of the COVID-19 pandemic in Saudi Arabia using ARIMA prediction model under current public health interventions [J].
Alzahrani, Saleh I. ;
Aljamaan, Ibrahim A. ;
Al-Fakih, Ebrahim A. .
JOURNAL OF INFECTION AND PUBLIC HEALTH, 2020, 13 (07) :914-919
[2]  
Anand A., LETT APPL NANOBIOSCI, V10, P2044
[3]   The proximal origin of SARS-CoV-2 [J].
Andersen, Kristian G. ;
Rambaut, Andrew ;
Lipkin, W. Ian ;
Holmes, Edward C. ;
Garry, Robert F. .
NATURE MEDICINE, 2020, 26 (04) :450-452
[4]   How will country-based mitigation measures influence the course of the COVID-19 epidemic? [J].
Anderson, Roy M. ;
Heesterbeek, Hans ;
Klinkenberg, Don ;
Hollingsworth, T. Deirdre .
LANCET, 2020, 395 (10228) :931-934
[5]  
[Anonymous], 2020, LANCET, V395, P1315, DOI 10.1016/S0140-6736(20)30938-7
[6]  
[Anonymous], World Economic Outlook, April 2020: The Great Lockdown
[7]  
[Anonymous], 2020, COVID 19 INDIA DATA
[8]   Forecast and evaluation of COVID-19 spreading in USA with re duce d-space Gaussian process regression [J].
Arias Velasquez, Ricardo Manuel ;
Mejia Lara, Jennifer Vanessa .
CHAOS SOLITONS & FRACTALS, 2020, 136
[9]  
Athreya S., 2020, COVID 19 INDIA TIMEL
[10]  
Atkeson A, 2020, National Bureau of Economic Research Working Paper no. 26867, DOI DOI 10.3386/W26867