Credit score: Pixabay/CC0 Public Area
A multidisciplinary group of researchers from the USC Faculty of Superior Computing and the Keck Faculty of Drugs, working alongside consultants from the Microsoft AI for Good Lab, Amref Well being Africa, and Kenya’s Ministry of Well being, has developed a man-made intelligence (AI) mannequin that may predict acute youngster malnutrition in Kenya as much as six months prematurely.
The instrument affords governments and humanitarian organizations important lead time to ship life-saving meals, well being care, and provides to at-risk areas. The machine studying mannequin outperforms conventional approaches by integrating scientific information from greater than 17,000 Kenyan well being amenities with satellite tv for pc information on crop well being and productiveness.
It achieves 89% accuracy when forecasting one month out, and maintains 86% accuracy over six months—a big enchancment over easier baseline fashions that rely solely on current historic youngster malnutrition prevalence developments.
In distinction to present fashions, the brand new instrument is very efficient at forecasting malnutrition in areas the place prevalence fluctuates and surges are tough to anticipate.
“This model is a game-changer,” stated Bistra Dilkina, affiliate professor of pc science and co-director of the USC Heart for Synthetic Intelligence in Society. “By using data-driven AI models, you can capture more complex relationships between multiple variables that work together to help us predict malnutrition prevalence more accurately.”
The findings are detailed in a PLOS One examine titled “Forecasting acute childhood malnutrition in Kenya using machine learning and diverse sets of indicators.”
The examine was co-authored by Girmaw Abebe Tadesse (Microsoft AI for Good Lab), Laura Ferguson (USC Institute on Inequalities in World Well being), Caleb Robinson, Rahul Dodhia, Juan M. Lavista Ferres (Microsoft AI for Good Lab), Shiphrah Kuria, Herbert Wanyonyi, Samuel Mburu (Amref Well being Africa), Samuel Murage (Kenyan Ministry of Well being), and Bistra Dilkina (USC Heart for AI in Society).
Girmaw Abebe Tadesse, principal scientist and supervisor on the Microsoft AI for Good Lab in Nairobi, Kenya, stated he believes the predictive AI instrument will make a distinction.
“This project is important, as malnutrition poses a significant challenge to children in Africa, a continent that is facing a major food insecurity exacerbated by climate change,” he stated.
A public well being emergency
In Kenya, 5% of youngsters underneath the age of 5—an estimated 350,000 people—undergo from acute malnutrition, a situation that weakens the immune system and dramatically will increase the danger of dying from frequent sicknesses like diarrhea and malaria. In some areas, the speed climbs as excessive as 25%. Globally, undernutrition is linked to almost half of all deaths in kids underneath 5.
“Malnutrition is a public health emergency in Kenya,” stated Laura Ferguson, director of analysis at USC’s Institute on Inequalities in World Well being and affiliate professor of inhabitants and public well being sciences on the Keck Faculty of Drugs of USC. “Children are sick unnecessarily. Children are dying unnecessarily.”
Present forecasting efforts in Kenya are largely based mostly on knowledgeable judgment and historic data—strategies that battle to anticipate new hotspots or speedy shifts.
As an alternative, the group’s mannequin makes use of Kenya’s routine well being information, collected via the District Well being Data System 2 (DHIS2), alongside satellite-derived indicators like crop well being and productiveness to establish rising threat areas with far better precision.
“The best way to predict the future is to create it using available data for better planning and prepositioning in developing countries,” stated Murage S.M. Kiongo, Program Officer for Monitoring and Analysis, Division of Diet and Dietetics, Ministry of Well being, Kenya. “Trends tell us a story. Multifaceted data sources, coupled with machine learning, offer an opportunity to improve programming on nutrition and health issues.”
The researchers have developed a prototype dashboard that visualizes regional malnutrition threat, enabling faster, better-targeted responses to youngster malnutrition dangers. Ferguson and Dilkina at the moment are working with the Kenyan Ministry of Well being and Amref Well being Africa to combine the mannequin and dashboard into authorities programs and resolution making, with the objective of making a sustainable and commonly up to date public useful resource.
“Most global health problems cannot be solved within the health field alone, and this is one of them,” Ferguson stated. “So, we absolutely need public health experts. We need medical officials. We need nonprofits. We need engineers. If you take out any single partner, it just doesn’t work and won’t have the impact that we hope for.”
Greater than 125 nations presently use DHIS2, together with about 80 low- and middle-income nations. Which means this AI-driven framework—which depends solely on present well being and satellite tv for pc information—may very well be tailored to combat malnutrition in different nations throughout the globe.
“If we can do this for Kenya, we can do it for other countries,” Dilkina stated. “The sky’s the limit when there is a genuine commitment to work in partnership.”
Extra info:
Girmaw Abebe Tadesse et al, Forecasting acute childhood malnutrition in Kenya utilizing machine studying and numerous units of indicators, PLOS One (2025). DOI: 10.1371/journal.pone.0322959
Offered by
College of Southern California
Quotation:
AI can predict youngster malnutrition and assist prevention efforts (2025, Might 14)
retrieved 14 Might 2025
from https://medicalxpress.com/information/2025-05-ai-child-malnutrition-efforts.html
This doc is topic to copyright. Aside from any honest dealing for the aim of personal examine or analysis, no
half could also be reproduced with out the written permission. The content material is offered for info functions solely.
