Wearable-derived joint kinematics for sleep posture analysis

Omar Elnaggar, Roselina Arelhi, Frans Coenen, Andrew Hopkinson, and Paolo Paoletti

Omar Elnaggar
School of Engineering and Materials Science, Queen Mary University of London, UK
Department of Electronic and Electrical Engineering, University College London, UK
School of Engineering, University of Liverpool, UK

Roselina Arelhi
Department of Electronic and Electrical Engineering, University College London, UK

Frans Coenen
School of Electrical Engineering, Electronics and Computer Science, University of Liverpool, UK

Andrew Hopkinson
School of Psychology, University of Liverpool, UK

Paolo Paoletti
School of Engineering, University of Liverpool, UK

Eighth UK Mobile, Wearable and Ubiquitous Systems Research Symposium (MobiUK 2026), 2026, pp. 1–1.

Abstract

Prolonged sleep in mechanically provocative postures may contribute to musculoskeletal symptoms, motivating unobtrusive methods for monitoring clinically relevant sleep posture. This work evaluates wearable-sensor-derived joint kinematics as biomechanical markers for classifying twelve complex sleep postures selected with clinical input. A body sensor network of eight magneto-inertial measurement units is combined with sensor fusion and a one-shot learning framework. Four joint orientations at the wrists and ankles provide posture signatures, while synthetic samples are generated from a single measured example of each posture. The framework was evaluated through in-silico, controlled outdoor in-vivo, and realistic indoor in-vivo studies. The in-silico study achieved complete classification accuracy, while the controlled study reliably distinguished approximately eleven postures at around 90% accuracy. Indoor testing was more challenging because of orientation drift, but one-shot augmentation still improved the number of distinguishable non-standard postures.

Documents: PDF Official MobiUK PDF MobiUK 2026 Programme

Suggested citation

O. Elnaggar, R. Arelhi, F. Coenen, A. Hopkinson, and P. Paoletti, “Wearable-derived joint kinematics for sleep posture analysis,” Eighth UK Mobile, Wearable and Ubiquitous Systems Research Symposium (MobiUK 2026), pp. 1–1, 2026.

References

  1. D. Cary, A. Jacques, and K. Briffa, “Examining relationships between sleep posture, waking spinal symptoms and quality of sleep: A cross sectional study,” PLOS ONE, vol. 16, no. 11, e0260582, 2021. https://doi.org/10.1371/journal.pone.0260582
  2. O. Elnaggar, F. Coenen, A. Hopkinson, L. Mason, and P. Paoletti, “Sleep posture one-shot learning framework based on extremity joint kinematics: In-silico and in-vivo case studies,” Information Fusion, vol. 95, pp. 215–236, 2023. https://doi.org/10.1016/j.inffus.2023.02.003

Back to Omar Elnaggar's publications