Novel Case-Based Reasoning System for Public Health Emergencies

Area of Science:

• Public Health
• Infectious Disease Epidemiology
• Computational Intelligence

Background:

• Recent decades have seen numerous infectious disease outbreaks (e.g., influenza, Ebola, SARS, COVID-19), highlighting the need for improved public health safety and suppression measures.
• Effective emergency treatment generation and rapid analysis methods are crucial for mitigating pandemic severity and saving lives.

Purpose of the Study:

• To develop a novel case-based reasoning (CBR) system for analyzing and managing public health emergencies.
• To enhance the speed and accuracy of infectious disease outbreak response through advanced computational methods.

Main Methods:

• A three-phase approach was employed: 1) calculating similarity between current and historical cases using heterogeneous information, 2) filtering relevant cases via grey clustering analysis, and 3) optimizing solutions using a cuckoo search (CS) algorithm.

Main Results:

• The proposed CBR model demonstrated superior performance compared to existing methods, including CBR with particle swarm optimization (PSO) and differential evolution (DE).
• The CS algorithm proved efficient in adapting solutions for public health emergencies.

Conclusions:

• The novel CBR system improves case retrieval speed using grey clustering and enhances solution accuracy with CS algorithms.
• This research offers a valuable tool for government agencies and public health organizations (e.g., CDC) to implement fast, accurate biohazard prevention and control measures.