Learning from Large-Scale Wearable Device Data for Predicting Epidemics Trend of COVID-19

Learning from Large-Scale Wearable Device Data for Predicting Epidemics Trend of COVID-19

Hui Wang, Jia Li, Rui Zhou, Yanan LI, Yi Yu, Xiao Tang, Yuling Dong, Zi Meng, Wang Huang, Guangxin Sun, Guokang Zhu, Kongqiao Wang

The pandemics of COVID-19 triggered out an alarm on the public health surveillance. The popularity of wearable devices enables a new perspective for the precaution of the infectious diseases. In this study, we propose a framework, which is based on the heart rate and sleep data collected from the wearable devices, to predict the epidemic trend of COVID-19 in different countries and cities. On top of a physiological anomaly detection algorithm defined based on wearable device data, an online neural network prediction modelling methodology combining both detected physiological anomaly rate and historical COVID-19 infection rate is explored. 4 models are trained separately according to geographical segmentation, i.e., North China, Central China, South China, and South-Central Europe. The de-identified sensor data from about 1.3 million wearable device users are used for verification. Experiment results indicate that the prediction models can be utilized to alert the outbreak of COVID-19 in advance, which sheds light on a health surveillance system with wearable device.

2020

DOI: 10.1155/2020/6152041

Discrete Dynamics in Nature and Society

Hindawi Limited

Mathematics