Are you getting sick? Predicting influenza-like symptoms using human mobility behaviors

Area of Science:

• Epidemiology
• Computational Social Science
• Public Health

Background:

• Individual mobility patterns are crucial for understanding and predicting the spread of infectious diseases.
• Personal mobility data can also offer insights into an individual's health status.
• Characterizing human movement is key for public health interventions.

Purpose of the Study:

• To investigate the impact of human mobility behaviors on predicting the onset of flu-like and cold symptoms.
• To develop and validate models for forecasting symptom presence using mobility data.

Main Methods:

• Utilized mobile phone mobility traces and daily self-reported symptoms from 29 individuals over a one-month period.
• Analyzed mobility trace characteristics such as total displacement, radius of gyration, and number of unique visited places.
• Developed predictive models based on individual mobility patterns.

Main Results:

• Demonstrated that daily symptoms can be predicted using an individual's mobility trace characteristics.
• Validated models capable of successfully predicting the future presence of flu-like and cold symptoms.
• Showcased the correlation between mobility patterns and symptom occurrence.

Conclusions:

• Mobility behavior analysis is a viable method for predicting individual health conditions, specifically flu-like and cold symptoms.
• The findings support the development of mobile applications for early disease detection and prevention.
• This approach has the potential to reduce disease spread and minimize contagion risks.