Hostname: page-component-cd9895bd7-gbm5v Total loading time: 0 Render date: 2024-12-28T19:32:59.807Z Has data issue: false hasContentIssue false
  • English
  • Français

HRNN4F: HYBRID DEEP RANDOM NEURAL NETWORK FOR MULTI-CHANNEL FALL ACTIVITY DETECTION

Published online by Cambridge University Press:  23 August 2019

Ahsen Tahir Open the ORCID record for Ahsen Tahir [Opens in a new window] ,
Jawad Ahmad ,
Gordon Morison ,
Hadi Larijani Open the ORCID record for Hadi Larijani [Opens in a new window] ,
Ryan M. Gibson  and
Dawn A. Skelton
Ahsen Tahir
Affiliation:
School of Computing, Engineering and Built Environment, Glasgow Caledonian University, Glasgow, UK Department of Electrical Engineering, University of Engineering and Technology, Lahore, Pakistan E-mails: [email protected]; [email protected]
Jawad Ahmad
Affiliation:
School of Computing, Engineering and Built Environment, Glasgow Caledonian University, Glasgow, UK
Gordon Morison
Affiliation:
School of Computing, Engineering and Built Environment, Glasgow Caledonian University, Glasgow, UK
Hadi Larijani
Affiliation:
School of Computing, Engineering and Built Environment, Glasgow Caledonian University, Glasgow, UK
Ryan M. Gibson
Affiliation:
School of Computing, Engineering and Built Environment, Glasgow Caledonian University, Glasgow, UK
Dawn A. Skelton
Affiliation:
School of Health and Life Sciences, Glasgow Caledonian University, Glasgow, UK
Get access Rights & Permissions [Opens in a new window]

Abstract

Falls are a major health concern in older adults. Falls lead to mortality, immobility and high costs to social and health care services. Early detection and classification of falls is imperative for timely and appropriate medical aid response. Traditional machine learning models have been explored for fall classification. While newly developed deep learning techniques have the ability to potentially extract high-level features from raw sensor data providing high accuracy and robustness to variations in sensor position, orientation and diversity of work environments that may skew traditional classification models. However, frequently used deep learning models like Convolutional Neural Networks (CNN) are computationally intensive. To the best of our knowledge, we present the first instance of a Hybrid Multichannel Random Neural Network (HMCRNN) architecture for fall detection and classification. The proposed architecture provides the highest accuracy of 92.23% with dropout regularization, compared to other deep learning implementations. The performance of the proposed technique is approximately comparable to a CNN yet requires only half the computation cost of the CNN-based implementation. Furthermore, the proposed HMCRNN architecture provides 34.12% improvement in accuracy on average than a Multilayer Perceptron.

Keywords

probabilistic networksqueueing theorysimulation
Type
Research Article
Information
Probability in the Engineering and Informational Sciences , Volume 35 , Special Issue 1: Learning, Optimization, and Theory of G-Networks , January 2021 , pp. 37 - 50
Copyright
Copyright © Cambridge University Press 2019

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1.Basterrech, S., Mohammed, S., Rubino, G., & Soliman, M. (2009). Levenberg–marquardt training algorithms for random neural networks. The Computer Journal 54: 125–135.CrossRefGoogle Scholar
2.Boyle, T. & Ravenscroft, A. (2012). Context and deep learning design. Computers and Education 59: 1224–1233.CrossRefGoogle Scholar
3.Brun, O., Yin, Y., & Gelenbe, E. (2018). Deep learning with dense random neural network for detecting attacks against IoT-connected home environments. Procedia Computer Science 134: 458–463.Google Scholar
4.Chen, Y. & Xue, Y. (2015). A deep learning approach to human activity recognition based on single accelerometer. In 2015 IEEE International Conference on Systems, Man, and Cybernetics, IEEE, pp. 1488–1492.CrossRefGoogle Scholar
5.Cleland, I., Kikhia, B., Nugent, C., Boytsov, A., Hallberg, J., Synnes, K., McClean, S., & Finlay, D. (2013). Optimal placement of accelerometers for the detection of everyday activities. Sensors 13: 9183–9200.CrossRefGoogle ScholarPubMed
6.Consultants PCP market research (2012). Falls : measuring the impact on older people. http://www.wrvs.org.uk/Uploads/Documents/Reports and Reviews/Falls report_web_v2.pdf.Google Scholar
7.Coskun, D., Incel, O.D., & Ozgovde, A. (2015). Phone position/placement detection using accelerometer: Impact on activity recognition. In 2015 IEEE Tenth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP), IEEE, pp. 1–6.CrossRefGoogle Scholar
8.Cramer, C.E. & Gelenbe, E. (2000). Video quality and traffic qos in learning-based subsampled and receiver-interpolated video sequences. IEEE Journal on Selected Areas in Communications 18: 150–167.CrossRefGoogle Scholar
9.Delaye, E., Sirasao, A., Dudha, C., & Das, S. (2017). Deep learning challenges and solutions with xilinx fpgas. In Proceedings of the 36th International Conference on Computer-Aided Design, IEEE, pp. 908–913.CrossRefGoogle Scholar
10.Gelenbe, E. (1989). Random neural networks with negative and positive signals and product form solution. Neural Computation 1: 502–510.CrossRefGoogle Scholar
11.Gelenbe, E. (1993). Learning in the recurrent random neural network. Neural Computation 5: 154–164.CrossRefGoogle Scholar
12.Gelenbe, E. & Hussain, K.F. (2002). Learning in the multiple class random neural network. IEEE Transactions on Neural Networks 13: 1257–1267.CrossRefGoogle ScholarPubMed
13.Gelenbe, E. & Stafylopatis, A. (1991). Global behavior of homogeneous random neural systems. Applied Mathematical Modelling 15: 534–541.CrossRefGoogle Scholar
14.Gelenbe, E. & Yin, Y. (2016). Deep learning with random neural networks. In Proceedings of SAI Intelligent Systems Conference, Springer, pp. 450–462.Google Scholar
15.Gelenbe, E. & Yin, Y. (2017). Deep learning with dense random neural networks. In International Conference on Man–Machine Interactions, Springer, pp. 3–18.Google Scholar
16.Gibson, R.M., Amira, A., Ramzan, N., Casaseca-de-la-Higuera, P., & Pervez, Z. (2016). Multiple comparator classifier framework for accelerometer-based fall detection and diagnostic. Applied Soft Computing 39: 94–103.CrossRefGoogle Scholar
17.Gibson, R.M., Amira, A., Ramzan, N., Casaseca-de-la Higuera, P., & Pervez, Z. (2017). Matching pursuit-based compressive sensing in a wearable biomedical accelerometer fall diagnosis device. Biomedical Signal Processing and Control 33: 96–108.CrossRefGoogle Scholar
18.Grenet, I., Yin, Y., Comet, J.-P., & Gelenbe, E. (2018). Machine learning to predict toxicity of compounds. In International Conference on Artificial Neural Networks, Springer, Cham, pp. 335–345.Google Scholar
19.Hinton, G. (2014). Lecture notes in neural networks for machine learning.Google Scholar
20.Khan, S.S. & Taati, B. (2017). Detecting unseen falls from wearable devices using channel-wise ensemble of autoencoders. Expert Systems with Applications 87: 280–290.CrossRefGoogle Scholar
21.Kwolek, B. & Kepski, M. (2014). Human fall detection on embedded platform using depth maps and wireless accelerometer. Computer Methods and Programs in Biomedicine 117: 489–501.CrossRefGoogle ScholarPubMed
22.Lane, N.D., Miluzzo, E., Lu, H., Peebles, D., Choudhury, T., & Campbell, A.T. (2010). A survey of mobile phone sensing. IEEE Communications Magazine 48: 140–150.CrossRefGoogle Scholar
23.Lu, N., Ren, X., Song, J., & Wu, Y. (2017). Visual guided deep learning scheme for fall detection. In 13th IEEE Conference on Automation Science and Engineering (CASE), 2017, IEEE, pp. 801–806.CrossRefGoogle Scholar
24.Martins, A. & Astudillo, R. (2016). From softmax to sparsemax: A sparse model of attention and multi-label classification. In International Conference on Machine Learning, pp. 1614–1623.Google Scholar
25.Masud, T. & Morris, R.O. (2001). Epidemiology of Falls. Age and Ageing 30(4): 3–7.CrossRefGoogle ScholarPubMed
26.Mathie, M.J., Coster, A.C., Lovell, N.H., & Celler, B.G. (2004). Accelerometry: providing an integrated, practical method for long-term, ambulatory monitoring of human movement. Physiological Measurement 25: R1.CrossRefGoogle ScholarPubMed
27.Nait Aicha, A., Englebienne, G., van Schooten, K.S., Pijnappels, M., & Kröse, B. (2018). Deep Learning to Predict Falls in Older Adults Based on Daily-Life Trunk Accelerometry. Sensors 18: 1–14.CrossRefGoogle ScholarPubMed
28.Noury, N., Rumeau, P., Bourke, A., ÓLaighin, G., & Lundy, J. (2008). A proposal for the classification and evaluation of fall detectors. Innovation and Research in Biomedical engineering (IRBM) 29: 340–349.Google Scholar
29.Ordóñez, F.J. & Roggen, D. (2016). Deep convolutional and lstm recurrent neural networks for multimodal wearable activity recognition. Sensors 16: 115.Google ScholarPubMed
30.Quero, J., Burns, M., Razzaq, M., Nugent, C., & Espinilla, M. (2018). Detection of falls from non-invasive thermal vision sensors using convolutional neural networks. In Multidisciplinary Digital Publishing Institute Proceedings, volume 2 p. 1236.CrossRefGoogle Scholar
31.Riedmiller, M. & Braun, H. (1993). A direct adaptive method for faster backpropagation learning: The rprop algorithm. In Proceedings of the IEEE international conference on neural networks, IEEE, pp. 586–591.CrossRefGoogle Scholar
32.Robbins, H. & Monro, S. (1951). A stochastic approximation method. The annals of mathematical statistics, pp. 400–407.Google Scholar
33.Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., & Salakhutdinov, R. (2014). Dropout: a simple way to prevent neural networks from overfitting. The Journal of Machine Learning Research 15: 1929–1958.Google Scholar
34.Sun, F., Wang, C., Gong, L., Xu, C., Zhang, Y., Lu, Y., Li, X., & Zhou, X. (2017). A power-efficient accelerator for convolutional neural networks. In 2017 IEEE International Conference on Cluster Computing (CLUSTER), IEEE, pp. 631–632.CrossRefGoogle Scholar
35.Tian, Y., Thompson, J., Buck, D., & Sonola, L. (2013). Exploring the system-wide costs of falls in older people in Torbay. King's Fund.Google Scholar
36.Timotheou, S. (2010). The random neural network: A survey. The computer journal 53: 251–267.CrossRefGoogle Scholar
37.Wang, J., Chen, Y., Hao, S., Peng, X., & Hu, L. (2019). Deep learning for sensor-based activity recognition: A survey. Pattern Recognition Letters 119: 3–11.CrossRefGoogle Scholar
38.Xu, T., Zhou, Y., & Zhu, J. (2018). New advances and challenges of fall detection systems: A survey. Applied Sciences 8: 418.CrossRefGoogle Scholar
39.Xue-Wen, C. & Xiaotong, L. (2014). Big data deep learning: Challenges and perspectives. IEEE Access 2: 514–525.CrossRefGoogle Scholar
40.Yin, Y. & Gelenbe, E. (2016a). Deep learning in multi-layer architectures of dense nuclei. arXiv preprint arXiv:1609.07160.Google Scholar
41.Yin, Y. & Gelenbe, E. (2016b). Nonnegative autoencoder with simplified random neural network. arXiv preprint arXiv:1609.08151.Google Scholar
42.Yin, Y. & Gelenbe, E. (2017). Single-cell based random neural network for deep learning. In 2017 International Joint Conference on Neural Networks (IJCNN), IEEE, pp. 86–93.CrossRefGoogle Scholar
43.Zebin, T., Scully, P.J., & Ozanyan, K.B. (2017). Human activity recognition with inertial sensors using a deep learning approach. Proceedings of IEEE Sensors, pp. 1–3.Google Scholar
44.Zeiler, M.D. (2012). Adadelta: An adaptive learning rate method). arXiv preprint arXiv:1212.5701.Google Scholar