Constenla-Villoslada’s paper published in PNAS

Ph.D. student Susana Constenla-Villoslada is the lead author on a paper just published in the Proceedings of the National Academy of Sciences. Congrats Susana!

Significance

As the number and share of people suffering from food insecurity worldwide has risen over the past decade, the humanitarian response community increasingly seeks advances in early warning systems to target populations who need food assistance. Advances in Earth Observation data and in machine learning have excited interest in their potential to help with early warning and geographic targeting of food assistance. To date, however, the predictive performance of such models with respect to child acute malnutrition has disappointed. We show how predictive skill and predictors vary over time and demonstrate the high value of monthly monitoring of child anthropometry in sentinel sites. With such data it is feasible to generate reasonably accurate forecasts at time horizons of 6 mo.

Citation

Constenla-Villoslada, S., Liu,Y., McBride, L., Ouma, C., Mutanda, N., &  Barrett, C.B. (2025). High-frequency monitoring enables machine learning–based forecasting of acute child malnutrition for early warning, PNAS U.S.A. 122 (23) e2416161122, https://doi.org/10.1073/pnas.2416161122