At this year’s PETRA conference in July, we will have the opportunity to present insights from two of our studies in the domain of machine learning and physical activity. The first study concerns the topic of fatigue detection during exercise. In the second study, we introduce the tool „NeckWatcher“ to monitor a person’s sitting position in real time.

Title of Study 1 (Full Paper):

Small Data, Big Challenges: Pitfalls and Strategies for Machine Learning in Fatigue Detection

Authors

André Jeworutzki, Jan Schwarzer, Kai von Luck, Peer Stelldinger, Susanne Draheim, and Qi Wang

Abstract

This research addresses the pitfalls and strategies for machine learning with small data sets in the context of sensor-based fatigue detection. It is shown that many existing studies in this area rely on small data sets and that classification results can vary considerably depending on the evaluation method. Our analysis is based on a study with 46 subjects performing multiple sets of squat exercises in a laboratory setting. Data from ratings of perceived exertion, inertial measurement units, and pose estimation were used to train and compare different classifiers. Our findings suggest that commonly used evaluation methods, such as leave-one-subject-out, should be used with caution and may not lead to generalizable classifiers. Furthermore, challenges related to imbalanced data and oversampling are discussed.

Title of Study 2 (Poster Paper):

NeckWatcher: A Real-time Monitoring Tool for the Assessment of the Neck Posture

Authors

Iryna Trygub, Johanna Ahlf, Martina Campanale, André Jeworutzki, and Jan Schwarzer

Abstract

Persistant poor posture can lead to the development of neck pain. Many different solutions have been proposed to aid in neck posture control, but most of them require additional devices. In this study, we present NeckWatcher, a tool that builds on MediaPipe Pose and utilizes an integrated webcam to monitor a person’s neck posture in real-time. It is designed to be user-friendly and realized as a stand-alone solution. Our first results suggest that NeckWatcher can be a useful tool for improving the sitting posture and, by that, reducing the risk of developing neck pain.