Revolutionizing COVID-19 Diagnosis: Dr. Michael Pokojovy’s AI-Powered Chest X-ray System

On March 12th, 2024, Dr. Michael Pokojovy, an Associate Professor of Data Science and School of Data Science Statistics Fellow at Old Dominion University, Norfolk, VA, presented his seminar, “An Interpretable Multi-task System for Clinically Applicable COVID-19 Diagnosis using CXR,” at UNC Charlotte as a part of The Center for TAIMing AI’s seminar series. In this seminar, Dr. Pokojovy proposed a system for diagnosing COVID-19 through chest X-rays using deep learning models.

With the virus continuously mutating into new variants, fast and reliable detection is important. Dr. Pokojovy proposed a system that can assist medical professionals with early COVID-19 diagnosis. This system preprocesses chest X-rays and applies unsupervised machine learning for lung region detection. The model is then able to classify the chest X-rays into three categories: COVID-19, pneumonia, and normal cases.

Using the Grad-CAM technique, class-specific heatmap images are created to enhance the trustworthiness of the medical AI system. Dr. Pokojovy’s proposed system quality is comparable to that of a radiologist. By offering a reliable and interpretable system for COVID-19 diagnosis Dr. Pokojovy’s work is a significant leap forward in the medical industry as well as in the advancement of AI.