A pervasive and sensor-free Deep Learning system for Parkinsonian gait analysis

Parkinsonian gait is associated with life-threatening consequences such as fall risk in Parkinson patients. Conventional Parkinsonian gait analysis heavily relies on expensive sensors and human labor. In this work, we propose a sensor-free end-to-end system which enables the automated and accurate Parkinsonian gait detection and analysis upon the videos recorded by pervasive cameras. Specifically, we leverage Deep Learning technologies to extract the human skeleton in the video frame and address the camera random angle challenge. By analyzing the gait features, we train a classifier based on a binary decision tree. Out of 16 Parkinsonian gait and 13 healthy gait videos, our system is able to detect the Parkinsonian Gait with 93.75% accuracy and healthy gait with 100% accuracy.

[1]  Xiaogang Wang,et al.  Multi-source Deep Learning for Human Pose Estimation , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[2]  B. Eskofier,et al.  Sensor-based Gait Parameter Extraction with Deep Convolutional Neural Networks. , 2016, IEEE journal of biomedical and health informatics.

[3]  Meng Wang,et al.  Disease Inference from Health-Related Questions via Sparse Deep Learning , 2015, IEEE Transactions on Knowledge and Data Engineering.

[4]  Benoit M. Macq,et al.  Gait feature extraction in Parkinson's disease using low-cost accelerometers , 2011, 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[5]  Natalie E. Allen,et al.  Recurrent Falls in Parkinson's Disease: A Systematic Review , 2013, Parkinson's disease.

[6]  Robert Iansek,et al.  Characteristics of motor disturbance in Parkinson's disease and strategies for movement rehabilitation , 1996 .

[7]  Julius Hannink,et al.  Quantifying postural instability in Parkinsonian gait from inertial sensor data during standardised clinical gait tests , 2017, 2017 IEEE 14th International Conference on Wearable and Implantable Body Sensor Networks (BSN).

[8]  P. Olivier,et al.  Accuracy of the Microsoft Kinect sensor for measuring movement in people with Parkinson's disease. , 2014, Gait & posture.

[9]  Nooritawati Md Tahir,et al.  Parkinson Disease gait classification based on machine learning approach , 2012 .

[10]  Angelo Cangelosi,et al.  Head pose estimation in the wild using Convolutional Neural Networks and adaptive gradient methods , 2017, Pattern Recognit..

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

[12]  M. Breteler,et al.  Epidemiology of Parkinson's disease , 2006, The Lancet Neurology.

[13]  Catherine Dehollain,et al.  Gait assessment in Parkinson's disease: toward an ambulatory system for long-term monitoring , 2004, IEEE Transactions on Biomedical Engineering.

[14]  Ashraf Darwish,et al.  Wearable and Implantable Wireless Sensor Network Solutions for Healthcare Monitoring , 2011, Sensors.

[15]  Yaser Sheikh,et al.  OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  S. Fahn.  Unified Parkinson's Disease Rating Scale , 1987 .

[17]  V. Preedy,et al.  Unified Parkinson's Disease Rating Scale , 2010 .