Gesundheit zum Mitmachen is a longitudinal study conducted in Bad Schönborn every six years. In this study, participating residents complete a comprehensive test battery administered by sport scientists, covering psychological factors, physical performance, and coordination. In the current round of the study conducted in the spring of 2025, the OpenEarable 2.0 (Röddiger et al., 2025; Official OpenEarble Website) was used as one of the measurement devices. During a 2 km walking test, the device was worn to collect movement data (IMU) and photoplethysmography (PPG) signals directly from the earable. The central research question of this thesis is to explore whether the standalone data from the OpenEarable 2.0 can be used to predict (some of) the outcomes of the sport science tests of the overall test battery.

1. Data Linkage: Separate recordings were made using the study companion app, which stored participant IDs and heart rate data from a Polar H10 sensor. In parallel, the OpenEarable 2.0 collected 9-axis IMU data, PPG signals, and pressure data. Since the pressure signal was also transmitted to the app, it can be used to synchronize the two datasets. In addition, participant IDs allow linking the sport scientists’ test results to the OpenEarable 2.0 data. After this matching process, data for approximately 30 participants will be available for analysis.
2. Data Preprocessing: Inspect and clean the collected data to ensure quality and usability. As the dataset comes from naturalistic recordings, it may contain irregularities such as device drops, sensor noise, or recording errors, which need to be identified and handled appropriately.
3. Data Exploration: Analyze the relationships between the signals measured with the OpenEarable 2.0 and the outcomes of the sport tests. This should be done using a range of methodological approaches (e.g., time series analysis, feature-based methods, or deep learning techniques) in order to capture different aspects of the data. Importantly, the evaluation should not only focus on predictive accuracy but also on the explainability of the results, with the goal of understanding which characteristics of the signals are most informative and how they relate to the sport test outcomes. In addition, heart rate data recorded with the Polar belt can serve as a valuable reference for comparison and validation.
4. Bonus (Optional): Go beyond the provided dataset by conducting a confirmatory analysis: collect additional data with the OpenEarable 2.0 during a set of defined exercises in a new participant group, and compare these results to the original findings.

• Interest in data science and its application to real-world, naturalistic settings
• Ability to systematically analyze and resolve challenges in the data analysis pipeline
• Strong Python skills
• German Language skills the Gesundheit zum Mitmachen documentation is in German

• A paragraph explaining your motivation.
• Your study program (Bachelor/Master), current semester, and field of study.
• A transcript of records (courses and grades).
• Your programming experience.
• Any areas of interest relevant to the topic.
• Your CV (if available)