A United States-based Nigerian researcher, Oluwatobi Akinbode, has developed an artificial intelligence-powered tool designed to improve early detection of infectious disease outbreaks and strengthen public health surveillance.
He said the tool, PRISM-EW™ (Probabilistic Rt-Integrated Surveillance Model for Early Warning), uses AI and statistical modelling to analyse disease surveillance data, estimate transmission trends and provide early warnings of possible outbreaks.
Akinbode’s innovation is coming amid growing concerns over the threat infectious diseases pose to healthcare systems globally.
Public health experts have consistently emphasised the importance of timely outbreak detection, noting that integrating emerging technologies into surveillance could improve preparedness and reduce the health and economic consequences of epidemics.
Speaking with PUNCH Healthwise, Akinbode, who is a statistical AI researcher, said PRISM-EW™ was designed to strengthen infectious-disease forecasting, outbreak detection, and public-health decision-making.
According to him, the framework addresses a persistent challenge in disease surveillance: the difficulty of translating fragmented epidemiological data into timely, reliable intelligence.
Unlike conventional approaches that assess disease transmission, reported cases, and outbreak indicators independently, he said PRISM-EW™ integrates these functions into a unified statistical AI framework.
Akinbode said PRISM-EW™ five components – preprocess, reproduction, incidence, signal, and multi-signal fusion- combine surveillance-data processing, transmission modeling, probabilistic forecasting, and statistical outbreak detection.
Central to the framework, he explained, is the Outbreak Pressure Index, a scoring system that evaluates transmission trends, disease burden, abnormal case patterns, and forecast risks to classify emerging threats by severity.
“Public-health surveillance should go beyond documenting disease outbreaks after they occur. We need intelligent systems that can recognise early warning signals, quantify uncertainty, and support timely intervention,” Akinbode said.
The statistical AI researcher said his background in researchers shaped his approach in developing statistical tools that connect complex health datasets with practical decision-making.
“Disease prediction is not simply about generating forecasts. It is about understanding transmission patterns, measuring risk, and giving health authorities reliable evidence to act before conditions worsen,” Akinbode said.
He noted that his work draws on extensive experience in statistical modeling, AI, and population-health research.
At Duke University, Akinbode conducted quantitative research examining HIV-risk factors through network analysis and statistical methods. His work contributed to a reported 30 percent improvement in identifying individuals eligible for HIV pre-exposure prophylaxis.
He also conducted environmental-health research at North Carolina A&T State University, applying Bayesian statistical methods to investigate relationships between heavy-metal exposure and cardiovascular-disease risk using data from the National Institutes of Health’s All of Us Research Program.
The findings were published in the International Journal of Environmental Research and Public Health.
Beyond PRISM-EW™, Akinbode said his professional portfolio includes AI engineering at LexisNexis, NSF-supported statistical computing research, scholarly publications, and international technology judging.
He noted that he is also a recipient of competitive research awards and contributed to academic peer review, reflecting his involvement in statistical research and emerging technologies.
Speaking further, he said PRISM-EW™ is designed for public-health agencies, epidemiological researchers, and healthcare surveillance programmes seeking to strengthen preparedness against infectious diseases, including influenza, COVID-19, measles, and cholera.
“Its significance lies in bringing established statistical and epidemiological methods into a coordinated predictive system capable of supporting more consistent outbreak-risk assessment.
“Artificial intelligence becomes valuable in healthcare when its predictions are interpretable, scientifically grounded, and useful to the people making critical decisions,” Akinbode added.
Copyright PUNCH
All rights reserved. This material, and other digital content on this website, may not be reproduced, published, broadcast, rewritten or redistributed in whole or in part without prior express written permission from PUNCH.
Contact: health_wise@punchng.
