Pre-and-post impact fall detection based on support vector machines using inertial and barometric pressure data

Detección de caídas con máquinas de soporte vectorial identificando pre y post impacto con datos inerciales y presión barométrica

Authors

Keywords:

support vector machines, activity recognition, fall detection, wearable inertial sensors, pre-and-post impact (en).

Keywords:

máquinas de soporte vectorial, reconocimiento de actividades , detección de caídas, sensores llevables inerciales, pre-post impacto (es).

Downloads

Abstract (en)

Objective: This paper presents a novel real-time algorithm for fall detection, which contextualizes falls by identifying activities occurring both pre- and post-impact utilizing machine learning techniques and wearable sensors.

Methodology: The activities selected to contextualize fall events included standing, lying, walking, running, climbing stairs, and using the elevator. Data were collected using an inertial measurement unit and a barometric altimeter positioned on the participants’ lower backs. Thirteen healthy subjects were observed performing the activities and fall events were recorded from five healthy subjects. The proposed algorithm combines thresholding and cascade support vector machines (SVMs), whose robustness is enhanced by a verification process of the subject’s posture aimed at determining the occurrence of the fall more accurately.

Results: The performance of the algorithm was evaluated in terms of the hit rate (HT) both offline and in real-time. From the activities studied, stairs climbing proved to be the most challenging to detect, with an offline HT of 85% and an online HT of 76 %. The overall offline performance was superior, with an HT of 96 %, compared to the performance achieved online, an HT of 91 %; in both cases the fall detection HT was 100 %.

Conclusions: The algorithm can be used to recognize fall events occurring to any user, as it has the advantage of not needing prior adaptation due to the nonlinear nature of the SVMs. The cascade SVMs allow for using small sets of variables, leading to low computational cost and a suitable real-time implementation. These features, in addition to the posture verification process, make our algorithm suitable for activity recognition in non-laboratory environments.

Abstract (es)

Objetivo: Este articulo presenta un algoritmo para la detección de caídas en tiempo real, el cual se contextualiza mediante el reconocimiento de las actividades previas y posteriores al impacto de la caída, utilizando técnicas de aprendizaje automático y sensores llevables.

Metodología: Las actividades estudiadas para la contextualización de la caída incluyen estar de pie, caminar, correr, subir o bajar en escaleras, y desplazarse en ascensor. La recolección de datos se hizo utilizando una unidad de medición inercial y un barómetro ubicados en la parte baja de la espalda de los participantes. Trece voluntarios sanos fueron observados para el registro de actividades y cinco voluntarios sanos para el registro de las caídas. El algoritmo propuesto combina máquinas de vectores de soporte (SVM) en cascada y umbrales, cuya robustez es mejorada con un proceso de verificación de la postura del sujeto para determinar de manera más precisa la ocurrencia de la caída.

Resultados: El rendimiento del algoritmo fue evaluado en términos de la tasa de aciertos (TA) tanto offline como en tiempo real de detección de las actividades estudiadas. Subir o bajar escaleras represento la mayor dificultad de detección, con una TA del 85% offline y del 76% online. El rendimiento global offline fue superior, con una TA del 96 %, comparado con el rendimiento alcanzado online, representado por una TA del 91 %; en ambos casos la TA de detección de caídas fue del 100 %.

Conclusiones: El algoritmo puede utilizarse para reconocer eventos en cualquier usuario, es decir, tiene la ventaja de no requerir adaptación previa, dada la naturaleza no lineal de las SVM. Las SVM en cascada permiten el uso de pequeños conjuntos de variables, lo que genera un bajo costo computacional y una implementación adecuada en tiempo real. Estas características, además de la robustez incrementada con un proceso de verificación de postura con excelentes resultados en la detección de la caída, permiten que nuestro algoritmo sea adecuado para el reconocimiento de actividades en entornos fuera del laboratorio.

Author Biographies

Roberth Álvarez-Jiménez , Industrial University of Santander

Master, Electronic Engineer. Industrial University of Santander. Bucaramanga, Colombia

Edith Pulido Herrera, Universidad El Bosque

European Doctor in Engineering, University Specialist in Mobile Communications), Electrical Engineer. Teacher Owner, El Bosque University. Bucaramanga, Colombia.

Andrés F. Ruiz-Olaya , Universidad Antonio Nariño

Doctor, Electronic Engineer. Associate Professor, Antonio Nariño University. Bogota

Daniel Alfonso Sierra Bueno, Industrial University of Santander

Doctor in Biomedical Engineering, Master in Electrical Engineering, Electrical and Electronic Engineer. Full Professor, University Santander Industrial. Bucaramanga, Colombia.

References

Álvarez, R., Pulido, E., & Sierra, D. A. (2017). Climbing/descending stairs detection using inertial sensors and implementing PCA and a SVM classifier. IFMBE Proceedings, 60. https://doi.org/10.1007/978-981-10-4086-3_146

Bravo, J. (2017). Diseño y desarrollo de un prototipo para inferir ubicación mediante el algoritmo dead reckoning utilizando métodos de corrección de errores. [Master’s Thesis, Universidad Industrial de Santander], Bucaramanga, Colombia.

Igual, R., Medrano, C., & Plaza, I. (2013). Challenges, issues and trends in fall detection systems. BioMed Eng OnLine, 12(66). https://doi.org/10.1186/1475-925X-12-66

Inman V, Ralston HJ, y Todd F. (2006). Human locomotion. In Rose J. & Gamble J. G. (Eds.), Human walking. (pp. 1-18).Williams & Wilkins.

James, S. L., Lucchesi, L. R., Bisignano, C., Castle, C. D., Dingels, Z. v., Fox, J. T., . . . Murray, C. J. L. (2019). The global burden of falls: Global, regional and national estimates of morbidity and mortality from the Global Burden of Disease Study 2017. Injury Prevention, 26(32). https://doi.org/10.1136/injuryprev-2019-043286

Kangas, M., Konttila, A., Lindgren, P., Winblad, I. y Jamsa, T. (2008). Comparison of low-complexity fall detection algorithms for body attached accelerometers. Gait and Posture, 28(2). https://doi.org/doi:10.1016/j.gaitpost.2008.01.003

Kumar, P. (2013). An Inertial Sensing Module for Movement and Posture Monitoring for Assisted Living. [Master’s Thesis, Indian Institute of Technology], Bombay, Bombay, India.

Lopez-Yunez, A., Vasquez, D., Palacio, L. A., Tiwari, N., Suryadevara, V. K., Anandwala, M. y Rizkalla, M. (2014). A novel approach for high speed wireless pre-fall detection multisensory system. Midwest Symposium on Circuits and Systems. 857–859. https://doi.org/10.1109/MWSCAS.2014.6908550

Nooruddin, S., Islam, Md. M., Sharna, F. A., Alhetari, H. & Kabir, M. N. (2021). Sensor-based fall detection systems: a review. Journal of Ambient Intelligence and Humanized Computing, 1-17. https://doi.org/10.1007/S12652-021-03248-Z

Noury, N., Fleury, A. & Rumeau, P. (2007). Fall Detection – Principles and Methods. 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, (pp. 1663–1666), https://doi.org/10.1109/IEMBS.2007.4352627

Patel, M., Pavic, A. & Goodwin, V. A. (2020). Wearable inertial sensors to measure gait and posture characteristic differences in older adult fallers and non-fallers: A scoping review. Gait and Posture, 76, 110–121. https://doi.org/10.1016/j.gaitpost.2019.10.039

Pierleoni, P., Belli, A., Palma, L., Pellegrini, M., Pernini, L. & Valenti, S. (2015). A High Reliability Wearable Device for Elderly Fall Detection. IEEE Sensors Journal, 15(8), 4544–4553. https://doi.org/10.1109/JSEN.2015.2423562

Sabatini, A. M., Ligorio, G., Mannini, A., Genovese, V. & Pinna, L. (2016). Prior-to- and post-impact fall detection using inertial and barometric altimeter measurements. IEEE Transactions on NeuralSystems and Rehabilitation Engineering, 24(7), 774–783. https://doi.org/10.1109/TNSRE.2015.2460373

Sorvala, A., Alasaarela, E., Sorvoja, H. y Myllyla, R. (2012). A two-threshold fall detection algorithm for reducing false alarms. 6th International Symposium on Medical Information and Communication Technology, SMICT 2012. https://doi.org/10.1109/ISMICT.2012.6203028

Tinetti, M. E., Liu, W. L., & Claus, E. B. (1993). Predictors and Prognosis of Inability to Get Up After Falls Among Elderly Persons. JAMA: The Journal of the American Medical Association, 269(1). https://doi.org/10.1001/jama.1993.03500010075035

Wild, D., Nayak, U. S. L., & Isaacs, B. (1981). How dangerous are falls in old people at home? British Medical Journal (Clinical Research Ed.), 282(6260). https://doi.org/10.1136/bmj.282.6260.266

World Health Organization. (2007). WHO Global Report on Falls Prevention in Older Age. Community Health.

Yu, X., Qiu, H. & Xiong, S. (2020). A Novel Hybrid Deep Neural Network to Predict Pre-impact Fall for Older People Based on Wearable Inertial Sensors. Front. Bioeng. Biotechnol., 12. https://doi.org/10.3389/fbioe.2020.00063

Zhang, T.,Wang, J., Xu, L. & Liu, P. (2006). Fall detection by wearable sensor and one-class SVM algorithm. Lecture Notes in Control and Information Sciences. 345. https://doi.org/doi:10.1007/11816515_104

How to Cite

APA

Álvarez-Jiménez , R., Pulido Herrera, E., Ruiz-Olaya , A. F., and Sierra Bueno, D. A. (2024). Pre-and-post impact fall detection based on support vector machines using inertial and barometric pressure data. Tecnura, 28(79), 34–52. https://doi.org/10.14483/22487638.22066

ACM

[1]
Álvarez-Jiménez , R. et al. 2024. Pre-and-post impact fall detection based on support vector machines using inertial and barometric pressure data. Tecnura. 28, 79 (Oct. 2024), 34–52. DOI:https://doi.org/10.14483/22487638.22066.

ACS

(1)
Álvarez-Jiménez , R.; Pulido Herrera, E.; Ruiz-Olaya , A. F.; Sierra Bueno, D. A. Pre-and-post impact fall detection based on support vector machines using inertial and barometric pressure data. Tecnura 2024, 28, 34-52.

ABNT

ÁLVAREZ-JIMÉNEZ , Roberth; PULIDO HERRERA, Edith; RUIZ-OLAYA , Andrés F.; SIERRA BUENO, Daniel Alfonso. Pre-and-post impact fall detection based on support vector machines using inertial and barometric pressure data. Tecnura, [S. l.], v. 28, n. 79, p. 34–52, 2024. DOI: 10.14483/22487638.22066. Disponível em: https://revistas.udistrital.edu.co/index.php/Tecnura/article/view/22066. Acesso em: 8 nov. 2024.

Chicago

Álvarez-Jiménez , Roberth, Edith Pulido Herrera, Andrés F. Ruiz-Olaya, and Daniel Alfonso Sierra Bueno. 2024. “Pre-and-post impact fall detection based on support vector machines using inertial and barometric pressure data”. Tecnura 28 (79):34-52. https://doi.org/10.14483/22487638.22066.

Harvard

Álvarez-Jiménez , R. (2024) “Pre-and-post impact fall detection based on support vector machines using inertial and barometric pressure data”, Tecnura, 28(79), pp. 34–52. doi: 10.14483/22487638.22066.

IEEE

[1]
R. Álvarez-Jiménez, E. Pulido Herrera, A. F. Ruiz-Olaya, and D. A. Sierra Bueno, “Pre-and-post impact fall detection based on support vector machines using inertial and barometric pressure data”, Tecnura, vol. 28, no. 79, pp. 34–52, Oct. 2024.

MLA

Álvarez-Jiménez , Roberth, et al. “Pre-and-post impact fall detection based on support vector machines using inertial and barometric pressure data”. Tecnura, vol. 28, no. 79, Oct. 2024, pp. 34-52, doi:10.14483/22487638.22066.

Turabian

Álvarez-Jiménez , Roberth, Edith Pulido Herrera, Andrés F. Ruiz-Olaya, and Daniel Alfonso Sierra Bueno. “Pre-and-post impact fall detection based on support vector machines using inertial and barometric pressure data”. Tecnura 28, no. 79 (October 27, 2024): 34–52. Accessed November 8, 2024. https://revistas.udistrital.edu.co/index.php/Tecnura/article/view/22066.

Vancouver

1.
Álvarez-Jiménez R, Pulido Herrera E, Ruiz-Olaya AF, Sierra Bueno DA. Pre-and-post impact fall detection based on support vector machines using inertial and barometric pressure data. Tecnura [Internet]. 2024 Oct. 27 [cited 2024 Nov. 8];28(79):34-52. Available from: https://revistas.udistrital.edu.co/index.php/Tecnura/article/view/22066

Download Citation

Visitas

18

Dimensions


PlumX


Downloads

Download data is not yet available.
Loading...