Artificial intelligence (AI) is opening new possibilities in healthcare to support decision-making and detect potential disease outbreaks. In a recent episode of TropmedTalk, Anis Fuad, S.Ked., DEA., a researcher at the Center for Tropical Medicine at UGM, discussed how AI can strengthen pandemic preparedness through data-driven early warning systems, while highlighting the importance of responsible implementation in healthcare. 

One initiative highlighted was AI for Pandemic Preparedness (AI4PEP), a global research initiative exploring the use of AI to improve readiness for future pandemic threats. As part of this initiative, the Center for Tropical Medicine at UGM is developing ExplainDengue, a platform that combines historical dengue data from healthcare facilities with climate, weather, and rainfall data to predict changes in dengue risk and identify areas that may experience rising cases. By providing early risk assessments, the platform aims to help health authorities anticipate potential increases in cases and better prepare their responses. 

Beyond dengue, the research team is also exploring the use of large language models (LLMs) can analyze electronic medical records from hospital emergency departments. By examining patients’ main complaints and medical histories, these systems aim to identify symptom patterns that could serve as early signals of a potential outbreak. However, Anis emphasized that effective AI depends on accurate, complete, consistent, and regularly updated data. Continuous evaluation and real-world implementation are equally important to ensure that promising innovations can be used in practice rather than remaining confined to pilot projects.

The discussion also highlighted the importance of governance and human oversight in healthcare AI. Technologies developed overseas should not automatically be adopted in Indonesia without assessing their performance, regulatory status, data management practices, and suitability for local healthcare settings. Anis stressed that humans involvement is essential throughout AI development, testing, and verification. This approach aligns with the World Health Organization’s guidance on AI ethics, which emphasizes protecting human autonomy. AI should therefore be used responsibly to support human decision-making, rather than becoming a substitute for appropriate professional judgment.

Watch the full TropmedTalk episode on AI in Healthcare and Pandemic Preparedness here.