Do We Walk Differently at Home? A Context-Aware Gait Analysis System in Continuous Real-World Environments
Roth N, Wieland GP, Küderle A, Ullrich M, Gladow T, Marxreiter F, Klucken J, Eskofier B, Kluge F (2021)
Publication Type: Conference contribution, Conference Contribution
Publication year: 2021
Original Authors: Nils Roth, Georg P. Wieland, Arne Kuderle, Martin Ullrich, Till Gladow, Franz Marxreiter, Jochen Klucken, Bjoern M. Eskofier, Felix Kluge
DOI: 10.1109/EMBC46164.2021.9630378
Abstract
Driven by the advancements of wearable sensors and signal processing algorithms, studies on continuous real-world monitoring are of major interest in the field of clinical gait and motion analysis. While real-world studies enable a more detailed and realistic insight into various mobility parameters such as walking speed, confounding and environmental factors might skew those digital mobility outcomes (DMOs), making the interpretation of results challenging. To consider confounding factors, context information needs to be included in the analysis. In this work, we present a context-aware mobile gait analysis system that can distinguish between gait recorded at home and not at home based on Bluetooth proximity information. The system was evaluated on 9 healthy subjects and 6 Parkinsons disease (PD) patients. The classification of the at home/not at home context reached an average F1-score of 98.2 ± 3.2 %. A context-aware analysis of gait parameters revealed different walking bout length distributions between the two environmental conditions. Furthermore, a reduction of gait speed within the at home context compared to walking not at home of 8.9 ± 9.4 % and 8.7 ±5.9 % on average for healthy and PD subjects was found, respectively. Our results indicate the influence of the recording environment on DMOs and, therefore, emphasize the importance of context in the analysis of continuous motion data. Hence, the presented work contributes to a better understanding of confounding factors for future real-world studies.
Authors with CRIS profile
Nils Roth
Machine Learning and Data Analytics Lab
Arne Küderle
Machine Learning and Data Analytics Lab
Martin Ullrich
Machine Learning and Data Analytics Lab
Till Gladow
Department of Molecular Neurology
Franz Marxreiter
Department of Molecular Neurology
Björn Eskofier
Machine Learning and Data Analytics Lab
Felix Kluge
Machine Learning and Data Analytics Lab
Involved external institutions
Luxembourg (LU)How to cite
APA:
Roth, N., Wieland, G.P., Küderle, A., Ullrich, M., Gladow, T., Marxreiter, F.,... Kluge, F. (2021). Do We Walk Differently at Home? A Context-Aware Gait Analysis System in Continuous Real-World Environments. In Proceedings of the 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC).
MLA:
Roth, Nils, et al. "Do We Walk Differently at Home? A Context-Aware Gait Analysis System in Continuous Real-World Environments." Proceedings of the 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) 2021.
BibTeX: Download