Charles Okon Esu, a researcher at the department of Civil and Environmental Engineering at Pusan National University in South Korea, has been honored by Brain Korea 21 with the 2024 Best Paper Award for his innovative research into the health impacts of air pollution using machine learning.
He triumphed over 110 other top experts to win this award.
His groundbreaking paper titled “Machine learning-derived dose-response relationships considering interactions in mixtures: Applications to the oxidative potential of particulate matter”, published in the highly regarded Journal of Hazardous Materials (Volume 475, August 2024), introduces Feature Localized Intercept Transformed-Shapley Additive Explanations (FLIT-SHAP), a cutting-edge explainable machine learning tool that provides an unprecedented understanding of how air pollutants interact and affect human health.
Before receiving this prestigious award, Charles’ paper had already garnered global recognition, with its findings featured in over 65 news articles worldwide. The paper’s transformative insights have redefined how researchers and policymakers approach air pollution, offering new ways to address its risks and protect public health.
Traditional air pollution studies typically focus on the effects of single pollutants in isolation. However, in the real world, people are exposed to complex mixtures of pollutants that interact in unexpected ways. FLIT-SHAP untangles these interactions, revealing which pollutants pose the greatest risks and how they combine to impact air quality.
Charles’ research unveiled significant findings. Combining laboratory simulated data and field measurements, pollutants mixtures significantly amplify or reduce the overall oxidative stress compared to summing each pollutant effect in the mixture. These results challenge conventional additive methods of evaluating air pollution health effect, demonstrating that the combined effects of pollutants are far more complex than previously understood. One of the study’s most important discoveries was identifying specific overlooked pollutants as significant contributors to health risks. This insight underscores the need to update current regulatory frameworks and prioritize pollutants with the most harmful effects.
Building on the success of this work, Charles Okon Esu is now developing and applying advanced machine learning tools to further explore the relationship between air quality, climate variables and health effects on a global scale. His research aims to develop data driven approaches that can extract information form air quality data or used for predicting air quality in data poor regions, enabling policymakers to implement more effective interventions and improve the quality of life for communities worldwide.
In the words of Charles Okon Esu:
“Winning this award is an incredible honor and reflects the significance of the tool developed in tackling real-world challenges. This research not only enhances our understanding of air pollution but also offers actionable solutions for creating cleaner air and healthier populations. I am deeply grateful for the support of my research team led by Prof. Cho Kuk whose guidance has been invaluable, the National Research Foundation of Korea and Brain Korea project, which provides an enabling environment for effective research.”
According to Charles, FLIT-SHAP’s applications can be extended beyond air quality research and have the potential to revolutionize several industries. Its advanced ability to analyze complex systems makes it highly applicable to areas such as agriculture, finance, and economics. In agriculture, for instance, FLIT-SHAP can be used to assess how various environmental factors, such as soil quality, weather patterns, and water availability, interact to influence crop yields. By understanding these complex relationships, farmers can optimize their strategies for better crop production and sustainability. In the field of finance, the tool can be employed to analyze the intricate interactions between economic indicators, market trends, and individual investments, helping to predict future market movements and manage risk more effectively. The versatility of FLIT-SHAP as an analytical tool makes it a powerful resource for tackling some of the most complex problems faced by industries worldwide.
On Thursday 23rd January 2025, Pusan National University celebrated Charles’ achievement along with other researchers as a significant milestone for its research community. His work exemplifies how cutting-edge multi-disciplinary research can address some of the world’s most pressing challenges.
Charles Okon Esu is open to collaborations and can be reached via email at esucharles286@gmail.com
