Optimal non-pharmaceutical intervention policy for Covid-19 epidemic via neuroevolution algorithm.

Area of Science:

• Epidemiology
• Public Health Policy
• Computational Biology

Background:

• COVID-19 pandemic responses varied globally, impacting economies.
• Epidemic control policies aim to disrupt transmission and prevent healthcare system overload, but incur economic costs.

Purpose of the Study:

• To develop an optimized intervention policy model for epidemic control.
• To identify strategies that minimize economic impact while protecting healthcare capacity.

Main Methods:

• Developed an intervention policy model incorporating human, implementation, and healthcare costs.
• Utilized a neuroevolution algorithm to identify optimal epidemic control strategies.
• Model determines minimum transmission reduction to keep healthcare burden below capacity.

Main Results:

• Optimal policy shows increased control strength early in the epidemic, followed by gradual decrease.
• The model provides adaptive intervention strategies without needing complete historical epidemic data.
• Demonstrates efficient policy adjustments at different epidemic stages.

Conclusions:

• Emphasizes the critical importance of early intervention measures.
• Provides insights into adaptive policies for minimizing economic consequences of epidemics.
• Highlights the potential to manage epidemics without exceeding healthcare system capacity.