Follow Us on Google
By Rita Okoye
In a world increasingly shaped by data-driven solutions, Gbenga Patrick Fabusola is pushing the frontiers of how artificial intelligence can reimagine the design and deployment of environmental sensors.
A Nigerian PhD candidate in Chemical Engineering with a minor in Artificial Intelligence at Oregon State University, Fabusola is leading groundbreaking research on the use of machine learning to accelerate the discovery of high-performance sensor materials—an innovation with vast implications for environmental monitoring, industrial safety, and smart agriculture.
At the heart of his work is a deep-learning-powered recommendation system inspired by the same algorithms behind Netflix or TikTok. But instead of recommending movies or music, Fabusola’s system identifies the best combinations of sensor materials—specifically metal-organic frameworks (MOFs)—to detect gases in various real-world scenarios.
“Sensor design is traditionally a slow, trial-and-error process,” he explained. “With over 100,000 MOFs available, testing each experimentally is not feasible. So we built an intelligent model that ranks the top-K sensors for specific applications, saving months—if not years—of guesswork.”
His approach has opened up new pathways for deploying gas sensors that can detect industrial leaks, monitor air quality, or even assist in medical diagnostics. These optimized sensors are capable of identifying volatile compounds at incredibly low concentrations, making them useful in detecting pollution, respiratory diseases, and chemical hazards in real time.
Fabusola’s research intersects materials science, data science, and chemical engineering. By simulating sensor responses using AI, his models not only predict how sensors will behave under different conditions but also help design entirely new sensor architectures optimized for sensitivity, selectivity, and speed.
“We are building a new paradigm where environmental sensors are smarter, faster, and tailored to the exact need,” he said. “This is especially important for resource-limited settings where every detection counts.”
His models have already been used to recommend sensor arrays for early gas leak detection in oil facilities and for monitoring indoor air pollutants—challenges particularly critical in Nigeria and other developing economies where industrial safety standards can lag behind global best practices.
Beyond environmental applications, Fabusola’s work is also finding relevance in bio-sensing. By applying gas sensing arrays to track bacterial and fungal emissions, he has demonstrated the potential for AI-enhanced systems to support healthcare diagnostics and pathogen surveillance.
A regular presenter at international conferences, including the American Institute of Chemical Engineers (AIChE), Fabusola has co-authored multiple peer-reviewed publications and is fast emerging as a thought leader in intelligent sensor technologies.
Despite the global relevance of his work, Fabusola’s long-term goals remain rooted in Africa. “Africa’s industrial and environmental challenges are complex, but solvable,” he emphasized. “With intelligent systems and predictive technologies, we can leapfrog many of the infrastructural constraints.”
Looking ahead, he envisions a future where AI-driven sensor platforms are deployed across cities, farms, and factories in Africa—quietly monitoring, learning, and responding to environmental and safety threats before they escalate.
“In a world of rising complexity, we need smarter tools,” Fabusola said. “And that’s exactly what we’re building.”
