TY - JOUR
T1 - Hidden Markov Model based stride segmentation on unsupervised free-living gait data in Parkinson’s disease patients
AU - Roth, Nils
AU - Küderle, Arne
AU - Ullrich, Martin
AU - Gladow, Till
AU - Marxreiter, Franz
AU - Klucken, Jochen
AU - Eskofier, Bjoern M.
AU - Kluge, Felix
N1 - Funding Information:
Open Access funding enabled and organized by Projekt DEAL. This work was supported by the Bavarian Ministry for Economy, Regional Development & Energy via the Medical Valley Award 2017 (FallRiskPD Project). B. M. Eskofier gratefully acknowledges the support of the German Research Foundation (DFG) within the framework of the Heisenberg professorship programme (Grant Number ES 434/8-1). F.K., M.U., A.K. and J.K. received funding from the IMI Mobilise-D project (Grant Agreement 820820).
Publisher Copyright:
© 2021, The Author(s).
PY - 2021/6/3
Y1 - 2021/6/3
N2 - Background: To objectively assess a patient’s gait, a robust identification of stride borders is one of the first steps in inertial sensor-based mobile gait analysis pipelines. While many different methods for stride segmentation have been presented in the literature, an out-of-lab evaluation of respective algorithms on free-living gait is still missing. Method: To address this issue, we present a comprehensive free-living evaluation dataset, including 146.574 semi-automatic labeled strides of 28 Parkinson’s Disease patients. This dataset was used to evaluate the segmentation performance of a new Hidden Markov Model (HMM) based stride segmentation approach compared to an available dynamic time warping (DTW) based method. Results: The proposed HMM achieved a mean F1-score of 92.1% and outperformed the DTW approach significantly. Further analysis revealed a dependency of segmentation performance to the number of strides within respective walking bouts. Shorter bouts (< 30 strides) resulted in worse performance, which could be related to more heterogeneous gait and an increased diversity of different stride types in short free-living walking bouts. In contrast, the HMM reached F1-scores of more than 96.2% for longer bouts (> 50 strides). Furthermore, we showed that an HMM, which was trained on at-lab data only, could be transferred to a free-living context with a negligible decrease in performance. Conclusion: The generalizability of the proposed HMM is a promising feature, as fully labeled free-living training data might not be available for many applications. To the best of our knowledge, this is the first evaluation of stride segmentation performance on a large scale free-living dataset. Our proposed HMM-based approach was able to address the increased complexity of free-living gait data, and thus will help to enable a robust assessment of stride parameters in future free-living gait analysis applications.
AB - Background: To objectively assess a patient’s gait, a robust identification of stride borders is one of the first steps in inertial sensor-based mobile gait analysis pipelines. While many different methods for stride segmentation have been presented in the literature, an out-of-lab evaluation of respective algorithms on free-living gait is still missing. Method: To address this issue, we present a comprehensive free-living evaluation dataset, including 146.574 semi-automatic labeled strides of 28 Parkinson’s Disease patients. This dataset was used to evaluate the segmentation performance of a new Hidden Markov Model (HMM) based stride segmentation approach compared to an available dynamic time warping (DTW) based method. Results: The proposed HMM achieved a mean F1-score of 92.1% and outperformed the DTW approach significantly. Further analysis revealed a dependency of segmentation performance to the number of strides within respective walking bouts. Shorter bouts (< 30 strides) resulted in worse performance, which could be related to more heterogeneous gait and an increased diversity of different stride types in short free-living walking bouts. In contrast, the HMM reached F1-scores of more than 96.2% for longer bouts (> 50 strides). Furthermore, we showed that an HMM, which was trained on at-lab data only, could be transferred to a free-living context with a negligible decrease in performance. Conclusion: The generalizability of the proposed HMM is a promising feature, as fully labeled free-living training data might not be available for many applications. To the best of our knowledge, this is the first evaluation of stride segmentation performance on a large scale free-living dataset. Our proposed HMM-based approach was able to address the increased complexity of free-living gait data, and thus will help to enable a robust assessment of stride parameters in future free-living gait analysis applications.
KW - HMM
KW - IMU
KW - Machine learning
KW - Mobile gait analysis
KW - Stride borders
KW - Wearable sensors
UR - http://www.scopus.com/inward/record.url?scp=85107134028&partnerID=8YFLogxK
U2 - 10.1186/s12984-021-00883-7
DO - 10.1186/s12984-021-00883-7
M3 - Article
C2 - 34082762
AN - SCOPUS:85107134028
SN - 1743-0003
VL - 18
SP - 93
JO - Journal of NeuroEngineering and Rehabilitation
JF - Journal of NeuroEngineering and Rehabilitation
IS - 1
M1 - 93
ER -