By Regina Stracqualursi

After coming to the United States in 2017 to pursue a master’s degree in computer science at Rensselaer Polytechnic Institute, Nkechinyere N. Agu ’19G, ’22 Ph.D., who was born and raised in Nigeria, witnessed the disparities in health care firsthand. “All of the countries in West Africa put together have fewer radiologists than Boston, Massachusetts,” she said, citing one example.

Upon the completion of her master’s degree, Agu decided to pursue a doctorate in computer science at Rensselaer and focused her research on addressing this problem. As part of the Artificial Intelligence Research Collaboration, an artificial intelligence (AI) research initiative between Rensselaer and IBM, she worked on a project to create a system that uses AI techniques to help doctors detect and diagnose diseases faster. “The AIRC allowed me to focus on what I was passionate about and carry out research that I believe will have impact,” she said.

Working directly with medical experts; her advisers, Rensselaer Professor James Hendler and Associate Professor Pingkun Yan; and her mentor, Mehdi Moradio, manager of Image Analysis and Machine Learning Research at the IBM Almaden Research Center, Agu created a system that could identify the key regions of a chest X-ray and identify if a disease was present in any of those regions. It could then make a diagnostic prediction based on both the anatomical information and patient history.

“The goal is to bridge the gap between developing countries and places like the U.S. If you have millions of people to one radiologist, there is a high chance there could be an error in that radiologist’s work,” Agu said. “But, if you have an AI system that looks over the data, the AI system could detect an error or a finding the radiologist missed.”

After earning her doctorate in August, Agu began working as an applied scientist for Amazon Web Services in its Machine Learning Solutions Lab, where she continues to pursue her passion for AI, providing machine learning solutions to clients. “AI is not a one-size-fits-all solution,” she said. “It’s a very valuable tool that will work hand in hand with people to achieve whatever goal they set out to achieve.”