Tensor Factorisation and Transfer Learning for Sleep Pose Detection

In this study, a novel hybrid tensor factorisation and deep learning approach has been proposed and implemented for sleep pose identification and classification of twelve different sleep postures. We have applied tensor factorisation to infrared (IR) images of 10 subjects to extract group-level data patterns, undertake dimensionality reduction and reduce occlusion for IR images. Pre-trained VGG-19 neural network has been used to predict the sleep poses under the blanket. Finally, we compared our results with those without the factorisation stage and with CNN network. Our new pose detection method outperformed the methods solely based on VGG-19 and 4-layer CNN network. The average accuracy for 10 volunteers increased from 78.1% and 75.4% to 86.0%.

[1]  Vangelis Metsis,et al.  Monitoring breathing activity and sleep patterns using multimodal non-invasive technologies , 2015, PETRA.

[2]  J. Leeuw,et al.  Principal component analysis of three-mode data by means of alternating least squares algorithms , 1980 .

[3]  Adrian Hilton,et al.  Sleep Posture Classification using a Convolutional Neural Network , 2018, 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[4]  Saeid Sanei,et al.  Multiview classification of brain data through tensor factorisation , 2015, 2015 IEEE 25th International Workshop on Machine Learning for Signal Processing (MLSP).

[5]  Tomomasa Sato,et al.  Sensor pillow system: monitoring respiration and body movement in sleep , 2000, Proceedings. 2000 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2000) (Cat. No.00CH37113).

[6]  Tamir Hazan,et al.  Non-negative tensor factorization with applications to statistics and computer vision , 2005, ICML.

[7]  Enamul Hoque,et al.  Monitoring body positions and movements during sleep using WISPs , 2010, Wireless Health.

[8]  Miad Faezipour,et al.  A Resource-Efficient Planning for Pressure Ulcer Prevention , 2012, IEEE Transactions on Information Technology in Biomedicine.

[9]  R. Harshman,et al.  PARAFAC: parallel factor analysis , 1994 .

[10]  Saeid Sanei,et al.  Improving time–frequency domain sleep EEG classification via singular spectrum analysis , 2016, Journal of Neuroscience Methods.

[11]  Saeid Sanei,et al.  Classification of awake, REM, and NREM from EEG via singular spectrum analysis , 2015, 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[12]  Rainer Stiefelhagen,et al.  Sleep position classification from a depth camera using Bed Aligned Maps , 2016, 2016 23rd International Conference on Pattern Recognition (ICPR).

[13]  Pierre Comon,et al.  Tensor Decompositions, State of the Art and Applications , 2002 .

[14]  Weimin Huang,et al.  Multimodal Sleeping Posture Classification , 2010, 2010 20th International Conference on Pattern Recognition.

[15]  Yi-Ping Hung,et al.  Sleep posture classification with multi-stream CNN using vertical distance map , 2018, 2018 International Workshop on Advanced Image Technology (IWAIT).

[16]  Shahram Payandeh,et al.  Toward study of features associated with natural sleep posture using a depth sensor , 2016, 2016 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE).

[17]  Tamara G. Kolda,et al.  Tensor Decompositions and Applications , 2009, SIAM Rev..

[18]  Min Hong,et al.  Sleep Monitoring System Using Kinect Sensor , 2015, Int. J. Distributed Sens. Networks.

[19]  Vangelis Metsis,et al.  Non-invasive analysis of sleep patterns via multimodal sensor input , 2012, Personal and Ubiquitous Computing.

[20]  M. Kryger,et al.  Sleep apnea and body position during sleep. , 1988, Sleep.

[21]  Ming-Sui Lee,et al.  Multiparameter Sleep Monitoring Using a Depth Camera , 2012, BIOSTEC.

[22]  Murat Boysan,et al.  Sleeping Position, Dream Emotions, and Subjective Sleep Quality , 2004 .