Bayesian classification and analysis of gait disorders using image and depth sensors of Microsoft Kinect

This paper presents a novel method of Bayesian gait recognition using Microsoft (MS) Kinect image and depth sensors and skeleton tracking in three-dimensional space. Although video sequences acquired by a complex camera system enable a very precise data analysis, it is possible to use much simpler technical devices to analyze video frames with sufficient accuracy for many applications. The use of the MS Kinect allows a simple 3-D modeling using its image and depth sensors for data acquisition, resulting in a matrix of 640 × 480 elements used for spatial modeling of a moving body. The experimental part of the paper is devoted to the study of three data sets: (i) 18 individuals with Parkinson's disease, (ii) 18 healthy age-matched controls, and (iii) 15 trained young individuals forming the second reference set. The proposed algorithm involves methods for the estimation of the average stride length and gait speed of individuals in these sets. Digital signal processing methods and Bayesian probability classification algorithms are then used for gait feature analysis to recognize individuals suspected of having Parkinson's disease. The results include the estimation of the characteristics of selected gait features for patients with Parkinson's disease and for individuals from the reference sets, presentation of decision boundaries, and comparison of classification efficiency for different features. The achieved accuracy of the probabilistic classification was 94.1%.

[1]  William Fitzgerald,et al.  A Bayesian approach to tracking multiple targets using sensor arrays and particle filters , 2002, IEEE Trans. Signal Process..

[2]  Anil C. Kokaram,et al.  Interpolation of missing data in image sequences , 1995, IEEE Trans. Image Process..

[3]  Mário Sarcinelli-Filho,et al.  The use of bayesian networks for heart beat classification. , 2010, Advances in experimental medicine and biology.

[4]  A. Prochazka,et al.  Multicamera Systems in the Moving Body Recognition , 2006, Proceedings ELMAR 2006.

[5]  David Barber,et al.  Bayesian reasoning and machine learning , 2012 .

[6]  Hui Li,et al.  Kinect Based Dynamic Hand Gesture Recognition Algorithm Research , 2012, 2012 4th International Conference on Intelligent Human-Machine Systems and Cybernetics.

[7]  Yi Li,et al.  Multi-scenario gesture recognition using Kinect , 2012, 2012 17th International Conference on Computer Games (CGAMES).

[8]  Simon Haykin,et al.  Neural Networks and Learning Machines , 2010 .

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

[10]  Julian Magarey,et al.  Wavelet Transforms in Image Processing , 1998 .

[11]  Claudia Linnhoff-Popien,et al.  Gait Recognition with Kinect , 2012 .

[12]  Hossein Mousavi Hondori,et al.  A Review on Technical and Clinical Impact of Microsoft Kinect on Physical Therapy and Rehabilitation , 2014, Journal of medical engineering.

[13]  Stepán Obdrzálek,et al.  Accuracy and robustness of Kinect pose estimation in the context of coaching of elderly population , 2012, 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[14]  Ales Procházka,et al.  Discrimination of axonal neuropathy using sensitivity and specificity statistical measures , 2014, Neural Computing and Applications.

[15]  Joan Lasenby,et al.  Using Geometric Algebra for Optical Motion Capture , 2001 .

[16]  Geoffrey I. Webb,et al.  Alleviating naive Bayes attribute independence assumption by attribute weighting , 2013, J. Mach. Learn. Res..

[17]  Sander Oude Elberink,et al.  Accuracy and Resolution of Kinect Depth Data for Indoor Mapping Applications , 2012, Sensors.

[18]  H. Hashimoto,et al.  Human motion tracking of mobile robot with Kinect 3D sensor , 2012, 2012 Proceedings of SICE Annual Conference (SICE).

[19]  Simon J. Godsill,et al.  Bayesian methods in signal and image processing , 1999 .

[20]  A Min Tjoa,et al.  Performance Comparison between Naïve Bayes, Decision Tree and k-Nearest Neighbor in Searching Alternative Design in an Energy Simulation Tool , 2013 .

[21]  K. Thirupal Reddy,et al.  Heart Disease Prediction System Using Naïve Bayes , 2012 .

[22]  Ercan E. Kuruoglu,et al.  Image separation using particle filters , 2007, Digit. Signal Process..

[23]  Tom Fawcett,et al.  An introduction to ROC analysis , 2006, Pattern Recognit. Lett..

[24]  Nasser M. Nasrabadi,et al.  Pattern Recognition and Machine Learning , 2006, Technometrics.

[25]  Yiping Yang,et al.  Real-time Hand Gesture Recognition from Depth Images Using Convex Shape Decomposition Method , 2014, J. Signal Process. Syst..

[26]  M. Markovic,et al.  Kinect in neurorehabilitation: Computer vision system for real time hand and object detection and distance estimation , 2012, 11th Symposium on Neural Network Applications in Electrical Engineering.

[27]  M. Milovanovic,et al.  New gait recognition method using Kinect stick figure and CBIR , 2012, 2012 20th Telecommunications Forum (TELFOR).

[28]  Alvin Harvey Kam,et al.  Unsupervised multiscale image segmentation , 1999, Proceedings 10th International Conference on Image Analysis and Processing.

[29]  Toby Sharp,et al.  Real-time human pose recognition in parts from single depth images , 2011, CVPR.

[30]  Miao Yu,et al.  An Online One Class Support Vector Machine-Based Person-Specific Fall Detection System for Monitoring an Elderly Individual in a Room Environment , 2013, IEEE Journal of Biomedical and Health Informatics.

[31]  M. Zweig,et al.  Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. , 1993, Clinical chemistry.

[32]  J. Hanley,et al.  The meaning and use of the area under a receiver operating characteristic (ROC) curve. , 1982, Radiology.

[33]  Remco R. Bouckaert.  Naive Bayes Classifiers That Perform Well with Continuous Variables , 2004, Australian Conference on Artificial Intelligence.

[34]  Tomás Pajdla,et al.  3D with Kinect , 2011, 2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops).

[35]  Ankit Chaudhary,et al.  Pre-consultation help necessity detection based on gait recognition , 2015, Signal Image Video Process..

[36]  Jing Xu,et al.  Development of an omni-directional 3D camera for robot navigation , 2012, 2012 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM).

[37]  Andrew W. Fitzgibbon,et al.  Real-time human pose recognition in parts from single depth images , 2011, CVPR 2011.

[38]  Zhengyou Zhang,et al.  Microsoft Kinect Sensor and Its Effect , 2012, IEEE Multim..

[39]  Bülent Sankur,et al.  Bayesian Separation of Images Modeled With MRFs Using MCMC , 2009, IEEE Transactions on Image Processing.

[40]  Jeffrey M. Hausdorff,et al.  Dual tasking, gait rhythmicity, and Parkinson's disease: Which aspects of gait are attention demanding? , 2005, The European journal of neuroscience.

[41]  Songthip Ounpraseuth,et al.  Estimating misclassification error: a closer look at cross-validation based methods , 2012, BMC Research Notes.

[42]  Ernesto Damiani,et al.  Temporal Denoising of Kinect Depth Data , 2012, 2012 Eighth International Conference on Signal Image Technology and Internet Based Systems.

[43]  Maher A. Sid-Ahmed,et al.  Investigating the Performance of Naive- Bayes Classifiers and K- Nearest Neighbor Classifiers , 2007 .

[44]  Ji Won Yoon,et al.  Bayesian inference for improved single molecule fluorescence tracking. , 2008, Biophysical journal.

[45]  Franz Pernkopf,et al.  Stochastic margin-based structure learning of Bayesian network classifiers , 2013, Pattern Recognit..

[46]  Linda Denehy,et al.  Validity of the Microsoft Kinect for assessment of postural control. , 2012, Gait & posture.

[47]  Harry Zhang,et al.  Exploring Conditions For The Optimality Of Naïve Bayes , 2005, Int. J. Pattern Recognit. Artif. Intell..

[48]  David G. Stork,et al.  Pattern Classification , 1973 .

[49]  S. Gilman,et al.  Diagnostic criteria for Parkinson disease. , 1999, Archives of neurology.

[50]  Kai Heinrich,et al.  Is the MS Kinect suitable for motion analysis? , 2012 .

[51]  Fabio Stella,et al.  Continuous time Bayesian network classifiers , 2012, J. Biomed. Informatics.

[52]  Jan Bürmann,et al.  Principal Component Analysis of gait in Parkinson's disease: relevance of gait velocity. , 2014, Gait & posture.

[53]  Joan Lasenby,et al.  Multiple Hypothesis Tracking for Automatic Optical Motion Capture , 2002, ECCV.

[54]  Lynn Rochester,et al.  Is gait variability reliable in older adults and Parkinson's disease? Towards an optimal testing protocol. , 2013, Gait & posture.

[55]  Vipin Chaudhary,et al.  Automated segmentation of stromal tissue in histology images using a voting Bayesian model , 2013, Signal Image Video Process..

[56]  K OrJ.  Numerical Bayesian methods applied to signal processing , 1996 .

[57]  Majid Ahmadi,et al.  Investigating the Performance of Naive- Bayes Classifiers and K- Nearest Neighbor Classifiers , 2010, 2007 International Conference on Convergence Information Technology (ICCIT 2007).

[58]  Simon J. Godsill,et al.  Variational and stochastic inference for Bayesian source separation , 2007, Digit. Signal Process..

[59]  Ales Procházka,et al.  The MS kinect image and depth sensors use for gait features detection , 2014, 2014 IEEE International Conference on Image Processing (ICIP).

[60]  Gian Luca Mariottini,et al.  Evaluating the accuracy of a mobile Kinect-based gait-monitoring system for fall prediction , 2013, PETRA '13.

[61]  Ran Gilad-Bachrach,et al.  Full body gait analysis with Kinect , 2012, 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[62]  Anil C. Kokaram,et al.  Detection of missing data in image sequences , 1995, IEEE Trans. Image Process..