Intelligent Early Warning and Response System Based on Health System Routine Data and Environment Data to Improve National Health Resilience.

Overview

In Indonesia, where dengue fever threatens millions each year, an AI-powered early warning system is transforming how communities and health authorities predict, prepare for, and respond to outbreaks before they spiral out of control.

Dengue fever remains one of Indonesia's most persistent public health challenges, with outbreaks driven by complex, shifting patterns of mosquito populations, rainfall, and urban conditions that are difficult for traditional surveillance systems to track in time. Healthcare facilities have historically struggled to collect, interpret, and act on disease data quickly enough to prevent widespread harm.

Developed in close collaboration with Indonesia's Ministry of Health, the system continuously collects data on mosquito populations, rainfall, weather conditions, and online medical consultations, using AI algorithms to detect patterns and generate accurate outbreak predictions. Large language models (LLMs) read and summarize electronic medical records, enabling faster, better-informed decisions by public health authorities and frontline practitioners. The result is a system that is not only technically robust but contextually grounded in Indonesia's local health priorities and governance structures.

By delivering timely alerts and actionable guidance directly to communities, the system empowers residents to prepare before outbreaks peak, while strengthening healthcare facilities' capacity to manage and respond to emerging threats. The project offers a scalable, government-aligned model for responsible AI deployment in low-resource health settings one that could significantly reduce dengue-related illness and death across Indonesia and serve as a blueprint for AI-driven disease surveillance across Southeast Asia.