A smartphone-based zero-effort method for mitigating epidemic propagation

Area of Science:

• Epidemiology
• Public Health Technology
• Mobile Computing

Background:

• Emerging infectious diseases like COVID-19 and SARS pose significant global health threats.
• Tracking mild or asymptomatic cases is challenging due to rapid spread and limited resources.
• Effective early warning systems are crucial for mitigating epidemic propagation.

Purpose of the Study:

• To develop a low-cost, real-time, smartphone-based epidemic early warning system.
• To identify individuals exposed to infectious diseases and assess quarantine needs.
• To mitigate the spread of epidemics without additional infrastructure.

Main Methods:

• Utilizing a hierarchical attention mechanism and temporal convolutional network to recognize epidemic-related voice activity.
• Estimating social distance between users via smartphone sensors.
• Combining Wi-Fi network logs and social distance to determine spatiotemporal contact and duration.
• Estimating infection risk based on vocal activity, social distance, and contact time.

Main Results:

• The proposed method effectively identifies spatiotemporal contacts and estimates infection risk.
• Experimental validation in typical scenarios confirms the method's efficacy.
• The system demonstrated feasibility for large-scale application and integration with existing control systems.

Conclusions:

• A smartphone-based zero-effort epidemic warning method can effectively mitigate epidemic propagation.
• The system's reliance on built-in smartphone sensors and absence of historical data facilitates broad adoption.
• This approach offers a scalable and cost-effective solution for real-time epidemic surveillance.