Converting Health Challenges into Predictive Solutions: Joseph Egbemhenghe on Transforming Public Health Using AI and Mathematics

In Lokoja, Kogi State, malaria, maternal deaths, and under-resourced hospitals have long plagued Kogi State’s health system. For Joseph Egbemhenghe, a scientist educated at Kogi State University, though, those challenges became the bedrock of a remarkable scientific odyssey. His expertise is in fractional-order mathematical modeling, and his research is branching out into artificial intelligence (AI)-based predictive medicine, which is redefining how developing areas can forecast and prevent disease.

This interview with Joseph tells us how his early work in Lokoja shaped his research path, how AI can alter the narrative of Kogi’s health sector, and why local data is central to solving global medical challenges in Kogi State. Fadekemi Ajakaiye reports

Kogi State has experienced major health crises in the past, from Malaria to maternal deaths. How did these realities impact your scientific choices?
Since growing up and studying in Kogi State, it was hard to deny the health issues around us. Malaria was practically a seasonal certainty. You would glimpse whole households falling ill with fever after heavy rains. Malaria continues to represent almost 27 percent of all illnesses here and more than 60 percent of children under five are affected, the Kogi State Ministry of Health said. In 2024, more than 40,000 cases of malaria among young children were recorded, with hundreds of very severe cases. That reality led me to wonder: Why are these numbers so steady year after year, despite all these interventions? I started to understand that the answer was not, as I had initially thought, the biology of the disease but our failure to predict and adapt. We tend to respond only after outbreaks happen. My motivation was that I was using mathematics to understand how malaria spreads in time, how treatment affects it, and how to best prevent it.

Your initial research at Kogi State University was based on fractional-order mathematical models for malaria and chlamydia. For readers who may not be familiar with that, what do we have here, and why is that so potent?

Traditional disease models assume the past does not affect the present, and we treat every infection event as unique, isolated. But in the real world, it’s not true. There is memory for infection, immunity, and behavior. A malaria outbreak now depends in part on rainfall patterns, human mobility, and how well we treat cases months in advance. This memory effect can be captured by fractional-order modeling of the system, which uses equations that model how the system “remembers” its history. When you take fractional calculus and apply it to the dynamics of malaria or chlamydia, you become way more accurate in your predictions about how diseases evolve, according to my research. That’s akin to changing a blurry picture over to a high-res one. We found that increased contact rates and a diminished treatment efficiency sharply increase prevalence, but improving therapy and vaccination dramatically lowers the transmission of infection. Those data were consistent with what health workers at Kogi were recording in real time.

You’ve since merged your mathematical background with artificial intelligence. What is AI doing in your current work role?

The task now is to get those mathematical models smart. We can even provide an array of datasets to the model with real epidemiological information, from rainfall and population density to reports out of the clinics, treatment results, and social media buzz about health. It is used to constantly adjust the parameters, monitoring diseases, adjusting how they react and how much control they can exert. I develop systems with libraries like Python, MATLAB and TensorFlow that mimic thousands of potential things. Consider coming up with predictions of malaria outbreak lines up weeks in the future to figure out what communities will need more mosquito nets or antimalarial medications. That is where AI comes in to play: real-time adaptability. The logic is the same for maternal health. Data and socio-economic factors can help predictive models to determine high-risk regions for maternal or neonatal complications. That may assist the Kogi State Primary Health Care Development Agency in investing in preventive and primary health-care resources ahead of emergencies.

The government of Kogi has just allocated ₦4 billion for malaria prevention and deployed over 5,000 health-data tools throughout hospitals. How will your research help to reinforce that?

These are excellent steps, but investment is not enough. I believe if we don’t have predictive intelligence, we continue to pursue the problem rather than outrun it. My models connect with those data systems to anticipate where outbreaks or financial bottlenecks will arise. For instance, AI could analyze data from local clinics to figure out which wards are facing an increase in fever cases. Health officials can get automated alerts, alter drug supplies or even dispatch mobile clinics to the area. Simultaneously, predictive analytics can help hospital administrators plan cost management, allowing them to stay financially solvent while increasing patient access. What we’re developing is an integrated health intelligence platform, one that integrates epidemiology, logistics, and finance. That could help Kogi transition from reactive care to proactive health care management.

How do you envision your proposed research reaching beyond Kogi State?

The diseases are different, but the principles are the same: Whether they’re malaria in Kogi, influenza in the U.S., or dengue in Southeast Asia, every epidemic is governed by mathematical rules. These AI-driven fractional models can be employed anywhere in order to predict disease spread and reduce it for public-health agencies. I’m now extending this work into a wider research project aimed at creating AI-driven health-modeling systems. The goal is to help agencies such as the CDC or NIH with adaptive simulation systems for predicting outbreaks and enhancing vaccine strategies. But the underpinnings of each idea are derived from studying Kogi’s health issues.

In retrospect, in what way do you think that your time at Kogi State University helped shape you into the scientist you now are?
The intellectual courage KSU offered me to see mathematics not as an abstraction, but as empathy, a vernacular to solve authentic human woes. Each formula I wrote had faces behind it: Mothers battling fever, children absent from school for malaria, doctors fighting for little in the way of medicine. Those experiences helped give me scientific roots. They taught me that innovation does not begin in fancy labs; it begins in communities that have answers. My journey from Kogi to cutting-edge AI research demonstrates that local problems can lead to global solutions.
The message of Egbemhenghe is simple and profound data can save lives if we learn to listen to it. His AI-based models are already being explored as a means of smarter health planning throughout Africa. For Kogi State, his work represents something more: evidence that the same place defined by health struggle can produce the science that brings them to a close.

As he stated:

“Kogi’s health story doesn’t have to be one of struggle. With all things mathematics, AI, and cooperation, it can become one of vision, invention, and hope.