AI Applications in Public Health: A Review of Epidemic Monitoring, Epidemiological Analysis, and Health Management

Abstract

This review explores AI’s applications in public health, focusing on epidemic monitoring, epidemiological analysis, and health management, alongside key challenges and future directions. In epidemic monitoring,AI enables early detection and prediction: social media data powers systems like WHO’s EARS (analyzing multilingual COVID-19 narratives with superior precision); AI processes news articles to spot outbreak signals (while addressing misinformation); mobility data analysis via GPT/GCNs improves disease spread forecasting; and anomaly detection (e.g., Siamese neural networks on ECG data) identifies unusual healthcare patterns signaling outbreaks. For epidemiological analysis, AI advances understanding of disease dynamics:Gaussian Mixture models cluster COVID-19 cases to reveal hotspots; causal inference techniques (aided by XAI) uncover disease-risk factor links; multi-factor AI models personalize risk stratification (e.g., HIV prevention, cardiac plaque assessment); and BRBFNs/neural networks model transmission (e.g., COVID-19, TB) to optimize controls. In health management, AI enhances care delivery: deep learning aids early diagnosis (e.g.,graph networks for cervical cancer, retinal analysis for glaucoma); AI integrates multi-omics/clinical data for personalized treatments (e.g., oncology biomarkers, stroke outcome prediction); RPM systems (sensors, voice chatbots) enable remote monitoring; and AI-driven platforms boost public health education (e.g., adolescent behavior interventions). Challenges include data privacy (needing robust cybersecurity), algorithmic bias (requiring diverse datasets/audits), and ethical concerns (upholding equity/transparency). Future directions in volve AI in drug development, workforce training, and fostering multidisciplinary collaboration to unlock AI’s full potential for equitable public health improvement.