Abstract
A key challenge in wearable sensor-based fall prediction is the fact that a fall event can often be performed in several different ways, with each consisting of its own configuration of poses and their spatio-temporal dependencies. Furthermore, to enable fall prevention of a person from imminent falls, precise predictions need to be achieved as far in advance as possible. This leads us to define a multi-channel temporal network, which explicitly characterizes the spatio-temporal relationships within a sensor channel as well as the interrelationships among channels by a combination representation of positional embedding and channel embedding to manage these unique fine-grained configurations among channels of a particular fall event. In addition, a transformer encoder is devised to exchange both inner-channel and inter-channel information in the encoder structure, and as a result, all local spatio-temporal dependencies are globally consistent. Empirical evaluations on two benchmark datasets and one in-house dataset suggest our model significantly outperforms the state-of-the-art methods. Our code is available at: https://github.com/passenger-820/MCTN.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Blunda, L.L., Gutiérrez-Madroñal, L., Wagner, M.F., Medina-Bulo, I.: A wearable fall detection system based on body area networks. IEEE Access 8, 193060–193074 (2020). https://doi.org/10.1109/ACCESS.2020.3032497
Challa, S.K., Kumar, A., Semwal, V.B.: A multibranch CNN-BiLSTM model for human activity recognition using wearable sensor data. Visual Comput. 38, 1–15 (2021). https://doi.org/10.1007/s00371-021-02283-3
Dirgová Luptáková, I., Kubovčík, M., Pospíchal, J.: Wearable sensor-based human activity recognition with transformer model. Sensors 22(5), 1911 (2022). https://doi.org/10.3390/s22051911
Hemmatpour, M., Ferrero, R., Gandino, F., Montrucchio, B., Rebaudengo, M.: Internet of Things for fall prediction and prevention. J. Comput. Methods Sci. Eng. 18(2), 511–518 (2018). https://doi.org/10.3233/JCM-180806
Howcroft, J., Kofman, J., Lemaire, E.D.: Prospective fall-risk prediction models for older adults based on wearable sensors. IEEE Trans. Neural Syst. Rehabil. Eng. 25(10), 1812–1820 (2017). https://doi.org/10.1109/TNSRE.2017.2687100
Jung, H., Koo, B., Kim, J., Kim, T., Nam, Y., Kim, Y.: Enhanced algorithm for the detection of preimpact fall for wearable airbags. Sensors 20(5), 1277 (2020). https://doi.org/10.3390/s20051277
Kim, T.H., Choi, A., Heo, H.M., Kim, H., Mun, J.H.: Acceleration magnitude at impact following loss of balance can be estimated using deep learning model. Sensors 20(21), 6126 (2020). https://doi.org/10.3390/s20216126
Kim, W., Son, B., Kim, I.: ViLT: vision-and-language transformer without convolution or region supervision (2021). https://doi.org/10.48550/arXiv.2102.03334
Kraft, D., Srinivasan, K., Bieber, G.: Deep learning based fall detection algorithms for embedded systems, smartwatches, and iot devices using accelerometers. Technologies 8(4), 72 (2020). https://doi.org/10.3390/technologies8040072
Liu, L., Hou, Y., He, J., Lungu, J., Dong, R.: An energy-efficient fall detection method based on FD-DNN for elderly people. Sensors 20(15), 4192 (2020). https://doi.org/10.3390/s20154192
Musci, M., De Martini, D., Blago, N., Facchinetti, T., Piastra, M.: Online fall detection using recurrent neural networks on smart wearable devices. IEEE Trans. Emerg. Topics Comput. 9(3), 1276–1289 (2021). https://doi.org/10.1109/TETC.2020.3027454
Palmerini, L., Klenk, J., Becker, C., Chiari, L.: Accelerometer-based fall detection using machine learning: training and testing on real-world falls. Sensors 20(22), 6479 (2020). https://doi.org/10.3390/s20226479
Saadeh, W., Butt, S.A., Altaf, M.A.B.: A patient-specific single sensor IoT-based wearable fall prediction and detection system. IEEE Trans. Neural Syst. Rehabil. Eng. 27(5), 995–1003 (2019). https://doi.org/10.1109/TNSRE.2019.2911602
Sucerquia, A., López, J.D., Vargas-Bonilla, J.F.: SisFall: a fall and movement dataset. Sensors 17(1), 198 (2017). https://doi.org/10.3390/s17010198
Triwiyanto, T., Pawana, I.P.A., Purnomo, M.H.: An improved performance of deep learning based on convolution neural network to classify the hand motion by evaluating hyper parameter. IEEE Trans. Neural Syst. Rehabil. Eng. 28(7), 1678–1688 (2020). https://doi.org/10.1109/TNSRE.2020.2999505
Vaswani, A., et al.: Attention is all you need (2017). https://doi.org/10.48550/arXiv.1706.03762
Vavoulas, G., Chatzaki, C., Malliotakis, T., Pediaditis, M., Tsiknakis, M.: The MobiAct dataset: recognition of activities of daily living using smartphones. In: International Conference on Information and Communication Technologies for Ageing Well and E-Health, vol. 2, pp. 143–151. SCITEPRESS (2016). https://doi.org/10.5220/0005792401430151
Acknowledgments
This work was supported by grants from the National Natural Science Foundation of China (grant no. 62377040), the Chongqing Graduate Research Innovation Project Funding (project no. CYS23101), the National Natural Science Foundation of China (grant nos. 61977012, 62207007).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Ethics declarations
Ethical Statement
Here is an ethical statement for a fall prediction experiment to ensure that data collection follows ethical principles:
– The privacy of the subjects is required to maintain confidentiality, and all data and information collected during the experiment will not be disclosed to the public or provided to non-experimental personnel.
– Before the experiment, we provided detailed explanations to the individuals participating in the experiment and obtained their informed consent, including informing them of the purpose, methods, and potential risks of the experiment.
– Participants in the experiment received appropriate protection and care, and risks and inconveniences during the experiment were minimized as much as possible. When conducting the experiment, we evaluated the impact of the experiment on the health and safety of the participants and took necessary measures to reduce these risks.
– After data collection is complete, we take full responsibility for data processing and analysis, ensuring that the various relationships within the data are clearly explained. Additionally, when publishing or using experimental data, we consider the sensitivity and privacy of the data and place the protection of participant privacy at the forefront.
We hereby confirm that we strictly adhered to the above-mentioned ethical principles in the fall prediction experiment and protected the rights and dignity of the subjects.
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Liu, J., Li, X., Liao, G., Wang, S., Liu, L. (2023). MCTN: A Multi-Channel Temporal Network for Wearable Fall Prediction. In: De Francisci Morales, G., Perlich, C., Ruchansky, N., Kourtellis, N., Baralis, E., Bonchi, F. (eds) Machine Learning and Knowledge Discovery in Databases: Applied Data Science and Demo Track. ECML PKDD 2023. Lecture Notes in Computer Science(), vol 14174. Springer, Cham. https://doi.org/10.1007/978-3-031-43427-3_24
Download citation
DOI: https://doi.org/10.1007/978-3-031-43427-3_24
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-43426-6
Online ISBN: 978-3-031-43427-3
eBook Packages: Computer ScienceComputer Science (R0)