Comparison of algorithms and classifiers for stride detection using wearables

Sensor-based systems for diagnosis or therapy support of motor dysfunctions need methodologies of automatically stride detection from movement sequences. In this proposal, we developed a stride detection system for daily life use. We compared mostly used algorithms min–max patterns, dynamic time warping, convolutional neural networks (CNN), and automatic framing using two data sets of 32 healthy and 28 Parkinson’s disease (PD) persons. We developed an insole with force and IMU sensors to record the gait data. The PD patients carried out the standardized time up and go test, and the healthy persons a daily life activities test (walking, sitting, standing, ascending and descending stairs). As an automatically stride detection process for daily life use, we propose a first stride detection using automatic framing, and after normalization and resampling data a CNN is used. A F1-score of 0.938 (recall 0.968, precision 0.910) for time up and go test and of 0.944 (recall 0.992, precision 0.901) for daily life activities test were obtained for CNN. Compared to the other detection methods, up to 6% F-measure improvement was shown.

[1]  J. Jankovic,et al.  Movement Disorder Society‐sponsored revision of the Unified Parkinson's Disease Rating Scale (MDS‐UPDRS): Scale presentation and clinimetric testing results , 2008, Movement disorders : official journal of the Movement Disorder Society.

[2]  Carlo Menon,et al.  Exploration of Gait Parameters Affecting the Accuracy of Force Myography-Based Gait Phase Detection* , 2018, 2018 7th IEEE International Conference on Biomedical Robotics and Biomechatronics (Biorob).

[3]  Carlos Manuel Travieso-González,et al.  Clustering of Human Gait with Parkinson's Disease by Using Dynamic Time Warping , 2018, 2018 IEEE International Work Conference on Bioinspired Intelligence (IWOBI).

[4]  Peter Peer,et al.  Transformation Based Walking Speed Normalization for Gait Recognition , 2013, KSII Trans. Internet Inf. Syst..

[5]  Jeffrey M. Hausdorff,et al.  Gait variability and basal ganglia disorders: Stride‐to‐stride variations of gait cycle timing in parkinson's disease and Huntington's disease , 1998, Movement disorders : official journal of the Movement Disorder Society.

[6]  Alper Yilmaz,et al.  A Deep Learning Strategy for Stride Detection , 2018, 2018 IEEE SENSORS.

[7]  Ioannis Pachoulakis,et al.  Exergames for Parkinson's Disease patients: The balloon goon game , 2016, 2016 International Conference on Telecommunications and Multimedia (TEMU).

[8]  Peter Peer,et al.  Frame–based classification for cross-speed gait recognition , 2017, Multimedia Tools and Applications.

[9]  Björn Eskofier,et al.  Stride Segmentation during Free Walk Movements Using Multi-Dimensional Subsequence Dynamic Time Warping on Inertial Sensor Data , 2015, Sensors.

[10]  Eamonn J. Keogh,et al.  Exact indexing of dynamic time warping , 2002, Knowledge and Information Systems.

[11]  Jang Sik Park,et al.  Gait analysis and identification based on joint information using RGB-depth camera , 2017, 2017 14th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON).

[12]  W. Teo,et al.  Parkinson's Disease and the Environment , 2019, Front. Neurol..

[13]  Chandra Prakash,et al.  Linear Prediction Model for Joint Movement of Lower Extremity , 2018, Advances in Intelligent Systems and Computing.

[14]  O. Mazumder,et al.  Assessment of Insole Based Gait Feature Variation with Progression of Parkinson's Disease , 2018, 2018 IEEE SENSORS.

[15]  Mihail Popescu,et al.  Estimating walking speed, stride length, and stride time using a passive floor based electronic scavenging system , 2017, 2017 IEEE Sensors Applications Symposium (SAS).

[16]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[17]  Hesham H. Ali,et al.  Machine Learning and Similarity Network Approaches to Support Automatic Classification of Parkinson's Diseases Using Accelerometer-based Gait Analysis , 2019, HICSS.

[18]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[19]  Bjoern M. Eskofier,et al.  Mobile Stride Length Estimation With Deep Convolutional Neural Networks , 2018, IEEE Journal of Biomedical and Health Informatics.

[20]  Claude Roux,et al.  Forensic gait analysis - Morphometric assessment from surveillance footage. , 2019, Forensic science international.

[21]  Vivienne Sze,et al.  Efficient Processing of Deep Neural Networks: A Tutorial and Survey , 2017, Proceedings of the IEEE.

[22]  Hritik Bansal,et al.  An improved sex-specific and age-dependent classification model for Parkinson's diagnosis using handwriting measurement , 2019, Comput. Methods Programs Biomed..

[23]  Jörg Frochte Maschinelles Lernen: Grundlagen und Algorithmen in Python , 2020 .

[24]  Dong Seog Han,et al.  ANN-based stride detection using smartphones for Pedestrian dead reckoning , 2018, 2018 IEEE International Conference on Consumer Electronics (ICCE).

[25]  Max A. Little,et al.  Technology in Parkinson's disease: Challenges and opportunities , 2016, Movement disorders : official journal of the Movement Disorder Society.

[26]  M. Hoehn,et al.  Parkinsonism , 1967, Neurology.

[27]  Valentina Agostini,et al.  Segmentation and Classification of Gait Cycles , 2014, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[28]  Fuhong Wang,et al.  Precise Displacement Estimation From Time-Differenced Carrier Phase to Improve PDR Performance , 2018, IEEE Sensors Journal.

[29]  Hu Ng,et al.  Human Identification Based on Extracted Gait Features , 2011 .

[30]  Peter Peer,et al.  Human Skeleton Model Based Dynamic Features for Walking Speed Invariant Gait Recognition , 2014 .

[31]  Takashi Watanabe,et al.  A study on IMU-based stride length estimation for motor disabled subjects: A comparison under different calculation methods of rotation matrix , 2018, 2018 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI).

[32]  Mirjana B. Popovic,et al.  Challenges of Stride Segmentation and Their Implementation for Impaired Gait , 2018, 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).