Forecasting influenza activity using machine-learned mobility map

Abstract

Human mobility is a primary driver of infectious disease spread. However, existing data is limited in availability, coverage, granularity, and timeliness. Data-driven forecasts of disease dynamics are crucial for decision-making by health officials and private citizens alike. In this work, we focus on a machine-learned anonymized mobility map (hereon referred to as AMM) aggregated over hundreds of millions of smartphones and evaluate its utility in forecasting epidemics. We factor AMM into a metapopulation model to retrospectively forecast influenza in the USA and Australia. We show that the AMM model performs on-par with those based on commuter surveys, which are sparsely available and expensive. We also compare it with gravity and radiation based models of mobility, and find that the radiation model’s performance is quite similar to AMM and commuter flows. Additionally, we demonstrate our model’s ability to predict disease spread even across state boundaries. Our work contributes towards developing timely infectious disease forecasting at a global scale using human mobility datasets expanding their applications in the area of infectious disease epidemiology.

Practical geospatial and sociodemographic predictors of human mobility

“…The Google Aggregated Mobility Research Dataset (GAMRD) represents one such form of passively collected data, aggregated over users who have turned on the Location History setting, which is off by default. The applicability of this data has been shown to be comparable with more traditional forms of mobility signals, such as travel history surveys 14 , 20 . While these data are likely biased in terms of wealth, gender, and urbanicity 21 , 22 , mobile penetration rate and smartphone ownership within sub-Saharan Africa have been steadily increasing over the decades, with nearly half a billion people currently subscribing to mobile services and an estimated 65% of the population having access to a smartphone device by 2025 22 .…”

“…Recent studies using the mobility data used in these analyses have demonstrated that mobility can robustly predict disease spread (or a reduction in disease spread), and are on par with more comprehensive travel surveys such as commuter surveys or GPS devices 7 , 14 , 20 . Regardless, these data can require extensive technical expertise to work with, and can be difficult to obtain due to privacy concerns and anonymity.…”

Using Mobile Phone Data to Estimate the Relationship between Population Flow and Influenza Infection Pathways

“…In the present study, the population flow information that is not directly related to influenza was analyzed, and it was compared to the influenza pandemic. There have recently been cases of using GPS to track several infectious diseases [ 9 , 10 , 11 , 30 , 31 ]. In these previous reports, authors made various attempts, such as using information from GPS to track the mobility of individual cases and assess trends in epidemics in the region.…”

“…With the recent spread of mobile phones and the accumulation of location information, one can now trace the fine movements of many individuals. Hence, infectious disease epidemic tracking and forecasting have been attempted using data based on mobile phone Global Positioning System (GPS) location information [ 7 , 8 , 9 , 10 , 11 ]. Research goals related to population flow and infectious disease transmission, such as tracking overall trends, identifying unknown individuals who may have been infected, and verifying models and concepts, differ.…”

A general method for estimating the prevalence of influenza-like-symptoms with Wikipedia data

“…In recent years, thanks to machine learning techniques, new methods have been developed to estimate the activity level of diseases by harnessing new data sources made available from the usage of internet services and big data. Researchers have tried for many years using unconventional sources of data to make predictions, for example, tweets [ 8 – 10 ], Google’s search keywords [ 11 , 12 ], Google Flu Trends [ 13 ] social media [ 14 ], Internet-based surveys [ 15 ], and mobility and telephony data [ 16 – 18 ]. Other works have employed multiple data sources to improve prediction and forecasting capabilities [ 19 ].…”