ETRI Journal

  • 제45권6호
  • /
  • Pages.1079-1089
  • /
  • 2023
  • /
  • 1225-6463(pISSN)
  • /
  • 2233-7326(eISSN)
한국전자통신연구원 (Electronics and Telecommunications Research Institute)
권호 표지
DOI QR코드

DOI QR Code

Activity recognition of stroke-affected people using wearable sensor

  • 투고 : 2022.07.26
  • 심사 : 2023.01.25
  • 발행 : 2023.12.10
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

Stroke is one of the leading causes of long-term disability worldwide, placing huge burdens on individuals and society. Further, automatic human activity recognition is a challenging task that is vital to the future of healthcare and physical therapy. Using a baseline long short-term memory recurrent neural network, this study provides a novel dataset of stretching, upward stretching, flinging motions, hand-to-mouth movements, swiping gestures, and pouring motions for improved model training and testing of stroke-affected patients. A MATLAB application is used to output textual and audible prediction results. A wearable sensor with a triaxial accelerometer is used to collect preprocessed real-time data. The model is trained with features extracted from the actual patient to recognize new actions, and the recognition accuracy provided by multiple datasets is compared based on the same baseline model. When training and testing using the new dataset, the baseline model shows recognition accuracy that is 11% higher than the Activity Daily Living dataset, 22% higher than the Activity Recognition Single Chest-Mounted Accelerometer dataset, and 10% higher than another real-world dataset.

키워드

참고문헌

  1. P. Casale, O. Pujol, and P. Radeva, Personalization and user verification in wearable systems using biometric walking patterns, Personal Ubiquitous Comput. 16 (2012), no. 5, 563-580. https://doi.org/10.1007/s00779-011-0415-z
  2. S. Sivapatham, R. Ramadoss, A. Kar, and B. Majhi, Monaural speech separation using GA-DNN integration scheme, Appl. Acoust. 160 (2020), 107140.
  3. D. Ravi, C. Wong, B. Lo, and G.-Z. Yang, A deep learning approach to on-node sensor data analytics for mobile or wearable devices, IEEE J. Biomed. Health Inform. 21 (2016), no. 1, 56-64.
  4. Q. Zhu, Z. Chen, and Y. C. Soh, A novel semisupervised deep learning method for human activity recognition, IEEE Trans. Industr. Inform. 15 (2018), no. 7, 3821-3830.
  5. R. Zhu, Z. Xiao, Y. Li, M. Yang, Y. Tan, L. Zhou, S. Lin, and H. Wen, Efficient human activity recognition solving the confusing activities via deep ensemble learning, IEEE Access 7 (2019), 75490-75499. https://doi.org/10.1109/ACCESS.2019.2922104
  6. M.-O. Mario, Human activity recognition based on single sensor square HV acceleration images and convolutional neural networks, IEEE Sensors J. 19 (2018), no. 4, 1487-1498. https://doi.org/10.1109/JSEN.2018.2882943
  7. W. Sousa Lima, E. Souto, K. El-Khatib, R. Jalali, and J. Gama, Human activity recognition using inertial sensors in a smartphone: An overview, Sensors 19 (2019), no. 14, 3213.
  8. T. Hur, J. Bang, T. Huynh-The, J. Lee, J.-I. Kim, and S. Lee, Iss2Image: A novel signal-encoding technique for CNNbased human activity recognition, Sensors 18 (2018), no. 11, 3910.
  9. H. Amroun, M. H. H. Temkit, and M. Ammi, Best feature for CNN classification of human activity using IoT network, (IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData), Exeter, UK), 2017, pp. 943-950.
  10. R. Zhu, Z. Xiao, M. Cheng, L. Zhou, B. Yan, S. Lin, and H. Wen, Deep ensemble learning for human activity recognition using smartphone, (IEEE 23rd International Conference on Digital Signal Processing (DSP), Shanghai, China), 2018, pp. 1-5.
  11. M. Alessandrini, G. Biagetti, P. Crippa, L. Falaschetti, and C. Turchetti, Recurrent neural network for human activity recognition in embedded systems using PPG and accelerometer data, Electronics 10 (2021), no. 14, 1715.
  12. L. Falaschetti, G. Biagetti, P. Crippa, M. Alessandrini, D. F. Giacomo, and C. Turchetti, A lightweight and accurate RNN in wearable embedded systems for human activity recognition, Intelligent decision technologies, Springer, 2022, pp. 459-468.
  13. A. Bensalah, J. Chen, A. Fornes, C. Carmona-Duarte, J. Llados, and M. A. Ferrer, Towards stroke patients' upper-limb automatic motor assessment using smartwatches, (Pattern recognition. ICPR International Workshops and Challenges, Virtual Event), 2021, pp. 476-489.
  14. H. E. M. Braakhuis, J. B. J. Bussmann, G. M. Ribbers, and M. A. M. Berger, Wearable activity monitoring in day-to-day stroke care: a promising tool but not widely used, Sensors 21 (2021), no. 12, 4066.
  15. B. Bruno, F. Mastrogiovanni, and A. Sgorbissa, A public domain dataset for ADL recognition using wrist-placed accelerometers, (The 23rd IEEE International Symposium on Robot and Human Interactive Communication, Edinburgh, UK), 2014, pp. 738-743.
  16. E. De-La-Hoz-Franco, P. Ariza-Colpas, J. M. Quero, and M. Espinilla, Sensor-based datasets for human activity recognition-a systematic review of literature, IEEE Access 6 (2018), 59192-59210. https://doi.org/10.1109/ACCESS.2018.2873502
  17. S. Sivapatham, A. Kar, and R. Ramadoss, Performance analysis of various training targets for improving speech quality and intelligibility, Appl. Acoust. 175 (2021), 107817.
  18. K. Ashwini, R. Amutha, R. Rajavel, and D. Anusha, Classification of daily human activities using wearable inertial sensor, (International Conference on Wireless Communications Signal Processing and Networking (WISPNET), Chennai, India), 2020, pp. 1-6.
  19. N. A. Capela, E. D. Lemaire, and N. Baddour, Feature selection for wearable smartphone-based human activity recognition with able bodied, elderly, and stroke patients, PloS One 10 (2015), no. 4, e0124414.
  20. M. K. O'Brien, N. Shawen, C. K. Mummidisetty, S. Kaur, X. Bo, C. Poellabauer, K. Kording, and A. Jayaraman, Activity recognition for persons with stroke using mobile phone technology: Toward improved performance in a home setting, J. Med. Int. Res. 19 (2017), no. 5, e7385.
  21. D. Garcia-Gonzalez, D. Rivero, E. Fernandez-Blanco, and M. R. Luaces, A public domain dataset for real-life human activity recognition using smartphone sensors, Sensors 20 (2020), no. 8, 2200.
  22. M. Ahmed, N. Mehmood, A. Nadeem, A. Mehmood, and K. Rizwan, Fall detection system for the elderly based on the classification of shimmer sensor prototype data, Healthcare Inform. Res. 23 (2017), no. 3, 147-158. https://doi.org/10.4258/hir.2017.23.3.147
  23. L. Venkataramana, S. G. Jacob, R. Ramadoss, D. Saisuma, D. Haritha, and K. Manoja, Improving classification accuracy of cancer types using parallel hybrid feature selection on microarray gene expression data, Genes Genomics 41 (2019), no. 11, 1301-1313. https://doi.org/10.1007/s13258-019-00859-x
  24. S. Ashry, R. Elbasiony, and W. Gomaa, An LSTM-based descriptor for human activities recognition using IMU sensors, (Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics, ICINCO, Vol. 1, Porto, Portugal), 2018, pp. 494-501.