Master of Science (MS), Wright State University, 2024, Computer Science

Occupationally-acquired infections impact thousands of healthcare workers (HCWs) in the U.S., with many cases preventable through proper use of personal protective equipment (PPE). This study seeks to develop a robust system to enhance PPE compliance and reduce infection risks among HCWs. The objectives of this thesis are twofold: (1) to create a hybrid machine learning model that combines object detection and keypoint detection to ensure correct donning and doffing of PPE, and (2) to design a real-time feedback system using LED indicators and a display interface to offer actionable guidance to HCWs during PPE usage. The goal is to optimize the ML model for accurate PPE detection and evaluate its performance in IoT and edge systems for real-time feedback, ensuring effective user interaction. This approach aims to promote safer healthcare environments by improving PPE compliance and minimizing exposure risks. The findings of this thesis demonstrate that the developed model is compact, secure, and capable of real-time performance, making it well-suited for IoMT frameworks.

Committee: Fathi Amsaad Ph.D. (Advisor); Hugh P. Salehi Ph.D. (Committee Member); Wen Zhang Ph.D. (Committee Member) Subjects: Computer Science; Medical Imaging