Research
My research focuses on analyzing multi‑modal sensor data from wearable devices collected in home and clinical settings, using signal processing techniques and machine learning approaches. The wearables include consumer‑grade smartwatches, continuous glucose monitoring (CGM) sensors, and inertial measurement units (IMUs) that capture activity and movement.
In our project Sensor-S, in collaboration with Charité and Data4Life, investigate the impact of such wearable consumer devices on patient engagement during post-stroke rehabilitation.
Beyond post‑stroke applications, I also analyze wearable and clinical data from other neurological conditions, including migraine and chronic inflammatory demyelinating polyneuropathy (CIDP), to explore digital markers that can support monitoring.
Focus: Multi-Modal Sensor Data, Wearables, Machine Learning with Time Series Data, Multi-Task Learning, Neurological Diseases
Teaching
Supervised Projects
Winter Term 2025/2026: Bachelor Project
Supervised Thesis
| Time | Name | Topic |
|---|---|---|
| 2026 (ongoing) | Peregin Wahle | Designing and Implementing an Interface for Adapting and Reusing ECG Preprocessing Pipelines |
Personal
Work Experience
- Nov 2025 – Present: Research Associate in Digital Health at HPI
- Dec 2024 – Sept 2025: Student Worker in Digital Health at HPI (Master Thesis)
- Sept 2024 – Feb 2025: Research Internship in Digital Health at HPI
- Jan 2024 – Aug 2024: Student Worker at Universität zu Lübeck, Institute for Medical Informatics, Medical Data Science
Education
- 2023 – 2025: M.Sc. Medical Informatics, Universität zu Lübeck
- 2020 – 2023: B.Sc. Medical Informatics, Universität zu Lübeck
Publications
Uhlig, A., Brandebusemeyer, C., Stolp, F., Hozhabr Pour, H., & Arnrich, B. (2025). Examining Software Developers’ Cognitive Load During Daily Activities with Wearables. Student Conference Proceedings, 1(1), 1936. https://doi.org/10.18416/SCP.2025.1936
Huang, X., Schmelter, F., Uhlig, A., Irshad, M. T., Nisar, M. A., Piet, A., … Grzegorzek, M. (2024). Comparison of feature learning methods for non-invasive interstitial glucose prediction using wearable sensors in healthy cohorts: a pilot study. In Intelligent Medicine. https://doi.org/10.1016/j.imed.2024.05.002