This study investigates the potential of integrating an efficient, automated, no-code machine learning classification model with Glo Germ fluorescent visualization to enhance hand hygiene training for nurse educators and infection preventionists. This approach aims to streamline training processes, eliminating biases stemming from human error during direct observation under ultraviolet light, and ultimately contributes to reducing healthcare-associated infections through effective hand hygiene training. Methods: This study utilized Google's Teachable Machine - a web-based graphical user interface tool designed for developing custom machine learning classification models without requiring specialized coding skills. Simulated contamination of hands was achieved using Glo Germ. A diverse training dataset was created with images of germ-contaminated and germ-free hands. The model was trained and evaluated using varying Glo Germ quantities. Results: The trained model exhibited a 100% confidence rating in classifying germ-contaminated hand surfaces and an average confidence rating of 94% for germ-free hands. Overall, the model achieved a 97% average confidence rating across the test dataset. Conclusions: This study illustrates the feasibility of using machine learning classification alongside Glo Germ fluorescent visualization for the real-time detection of germs on hand surfaces. The integration of these techniques presents an efficient and accessible approach to enhance hand hygiene training methodologies for nurse educators and infection preventionists by: (i) providing automated and immediate visual feedback on handwashing effectiveness, (ii) addressing inherent limitations associated with in-person monitoring such as bias, and (iii) providing no-code machine learning tool to healthcare educators and practitioners who may lack coding experience.

Hand hygiene glogerm machine learning teachable machine