Home / Current Issue / Paper 1706660
Machine Learning-Driven Fall Detection Using Wearable Sensors for Enhanced?Safety
Subject area: Science,Engineering and Technology · Area of research: Artificial Intelligence and Wearable Technology
Abstract
The goal of this study is to create an accurate and dependable fall detection system utilizing machine learning methods. By merging accelerometer measurements from wearable sensors that covered motion patterns related with falls and regular activities, a huge dataset was constructed. To extract useful characteristics from sensor data, feature engineering approaches were used. After the training of the model, upon testing an accuracy value of 86.25% was attained, alongside a recall value of 90.68%, precision value of 83.16% and a f1 score of 85.91%. Finally, this work proposes a novel approach to fall detection based on machine learning approaches. Our research shows significant progress in accurately detecting falls, outperforming existing threshold-based approaches. The purpose of building an effective fall detection system is to promote individual safety and well-being, particularly in healthcare settings. The proposed technique has enormous potential to transform how we respond to fall-related events and provide crucial help to the aging population and those who work jobs that may cause them to fall, such as construction workers. This opens up fascinating new possibilities for future study and real-world use of machine learning-based fall detection systems, which will directly touch people's lives.
Keywords
Accelerometer, Fall detection, Machine learning-based, Wearable sensors.
References
[1] Hoskin, A. (1998, April 01). Europe PMC. Retrieved from https://europepmc.org: https://europepmc.org/article/med/9592923.
[2] World Health Organization. (2023, March 30). Retrieved from https://www.who.int/news-room/fact-sheets/detail/falls#:~:text=A%20fall%20is%20defined%20as,though%20most%20are%20non%2Dfatal.
[3] Bourke, A. K. (2010). GamblEvaluation of waist-mounted tri-axial accelerometer-based fall-detection algorithms during scripted and continuous unscripted activities. Journal of Biomechanics, 3051-3057.
[4] Mahesh, B. (2018). Machine Learning Algorithms - A Review. International Journal of Science and Research (IJSR), 381-386.
[5] Bush, R. R. (1951). A mathematical model for simple learning. Psychological Review, 58(5),313–323.
[6] Robert Koch. (2023, March 2). https://www.clickworker.com/. Retrieved from clickworker.com: https://www.clickworker.com/customer-blog/history-of-machine-learning/
[7] Foote, K. (2021, December 3). dataversity. Retrieved from https://www.dataversity.net: https://www.dataversity.net/a-brief-history-of-machine-learning/
[8] Mahapatra, S. (2018, March 21). Towards Data Science. Retrieved from towardsdatascience.com: https://towardsdatascience.com/why-deep-learning-is-needed-over-traditional-machine-learning-1b6a99177063
[9] F. Hussain, F. H.-u.-H. (2019). Activity-Aware Fall Detection and Recognition Based on Wearable Sensors. IEEE Sensors Journal, vol. 19, no. 12, 4528-4536.
[10] Adrian Nunez-Marcos, G. A.-C. (2017). Vision-Based Fall Detection with Convolutional Neural Networks. Hindawi,Wireless Communications and Mobile Computing 2017(1), 1-16.
[11] Cheffena, M. (2016). Fall Detection Using Smartphone Audio Features. IEEE Journal of Biomedical and Health Informatics, vol. 20, no. 4, 1073-1080.
[12] Minvielle, L. M. (2017). Fall detection using smart floor sensor and supervised learning. 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). Jeju, Korea (South): IEEE.
[13] Tang, C.-H. O. (2015). Fall Detection Sensor System for the Elderly. International Journal of Advanced Computer Research.
[14] Su, Y. &. (2015). Fall detection in the elderly population by using wearable devices: A review. Sensors, 15(6), 11882-11910.
[15] Santoyo-Ramón, J. C.-G. (2018). Analysis of a smartphone-based architecture with multiple mobility sensors for fall detection with supervised learning. Sensors 18, 1155.
How to cite this paper
@article{1706660,
author = {Obi-Obuoha Abiamamela, Ngim N. Ewezu, Adjeroh E. Princewill, Akinwumi O. Aderonke},
title = {Machine Learning-Driven Fall Detection Using Wearable Sensors for Enhanced?Safety},
journal = {Iconic Research And Engineering Journals},
year = {2024},
volume = {8},
number = {6},
pages = {280-294},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1706660.pdf},
abstract = {The goal of this study is to create an accurate and dependable fall detection system utilizing machine learning methods. By merging accelerometer measurements from wearable sensors that covered motion patterns related with falls and regular activities, a huge dataset was constructed. To extract useful characteristics from sensor data, feature engineering approaches were used. After the training of the model, upon testing an accuracy value of 86.25% was attained, alongside a recall value of 90.68%, precision value of 83.16% and a f1 score of 85.91%. Finally, this work proposes a novel approach to fall detection based on machine learning approaches. Our research shows significant progress in accurately detecting falls, outperforming existing threshold-based approaches. The purpose of building an effective fall detection system is to promote individual safety and well-being, particularly in healthcare settings. The proposed technique has enormous potential to transform how we respond to fall-related events and provide crucial help to the aging population and those who work jobs that may cause them to fall, such as construction workers. This opens up fascinating new possibilities for future study and real-world use of machine learning-based fall detection systems, which will directly touch people's lives.},
keywords = {Accelerometer, Fall detection, Machine learning-based, Wearable sensors.},
month = {December},
}