- April 02, 2026
- By Renata Johnson
A NASA-funded research project led by a University of Maryland geographer is injecting artificial intelligence into disease forecasting efforts through the addition of a key partner.
The nonprofit Institute for Health Modeling and Climate Solutions (IMACS) joined the initiative directed by geographical sciences Professor and Chair Tatiana Loboda; the organization will contribute expertise in deep learning and multidisease forecasting to help predict malaria and dengue outbreaks across Indonesia.
Launched last year with $1.1 million in NASA funding, the three-year project integrates satellite-based environmental monitoring, health modeling and artificial intelligence to produce high-resolution forecasts that could help guide public health interventions.
“IMACS brings tremendous expertise in deep learning models, a key component that was previously missing from this project,” said Loboda, the project’s principal investigator. “We’re very fortunate to work with IMACS, which has a unique combination of technical expertise and knowledge of malaria and vector-borne diseases. I don’t know of another team that can do both.”
IMACS is bringing its AI-powered multidisease forecasting model SPECTRA to the project, which was included in a casebook from the World Health Organization on using AI to benefit health in the global south.
“Combined with NASA’s Earth observation information and UMD’s advanced modeling capabilities, we are on the cusp of developing a breakthrough tool,” said Kaushik Sarkar, director of IMACS and co-investigator of the project. ““If we are successful, the project will empower local health authorities with high-resolution data, making a real difference in the push to end these deadly diseases.”
The model is being developed for Indonesia, a large country with varied landscapes, diverse ecosystems and distinct microclimates — all of which complicate efforts to eliminate malaria and control other vector-borne diseases like dengue. The final stage of the project, if successful, will be handing the model over to the Indonesian government.