AUTHOR=Ataguba Grace , Orji Rita 

TITLE=Toward the design of persuasive systems for a healthy workplace: a real-time posture detection

JOURNAL=Frontiers in Big Data

VOLUME=Volume 7 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/big-data/articles/10.3389/fdata.2024.1359906

DOI=10.3389/fdata.2024.1359906

ISSN=2624-909X

ABSTRACT=Persuasive technologies, in connection with human factor engineering requirements for healthy workplaces, have played a significant role in ensuring a change in human behavior. Healthy workplaces suggest different best practices applicable to body posture, proximity to the computer system, movement, lighting conditions, computer system layout, and other significant psychological and cognitive aspects. In this study, we explored body posture in line with best and healthy practices that suggest how users should sit or stand in workplaces. We found that most unhealthy postures have a long-term impact on the lifestyle and health of computer users. Besides, people work long or short hours on the computer system and have become less conscious of essential best practices. Though most persuasive studies are now beginning to provide reminders to computer users to take regular breaks from their computer systems, little attention has been paid to making computers responsive to computer users’ unhealthy workplace practices based on their bad postures. Given the significance of deep learning models in real-time object detection, we employed these models to support real-time detection in response to the unhealthy practices of computer users. Hence, this paper provides a real-time posture detection framework based on two deep learning models: convolutional neural networks (CNN) and Yolo-V3. Results show that our YOLO-V3 model outperformed CNN model with a mean average precision of 92%. Based on this finding, we provide implications for integrating proximity detection and designing persuasive systems for a healthy workplace in the future.