Internet of Things (IoT) Technologies for HealthCare. 4th International Conference, HealthyIoT 2017, Angers, France, October 24-25, 2017, Proceedings

Research Article

An Efficient Design of a Machine Learning-Based Elderly Fall Detector

Download
187 downloads
  • @INPROCEEDINGS{10.1007/978-3-319-76213-5_5,
        author={L. Nguyen and M. Saleh and R. Bouquin Jeann\'{e}s},
        title={An Efficient Design of a Machine Learning-Based Elderly Fall Detector},
        proceedings={Internet of Things (IoT) Technologies for HealthCare. 4th International Conference, HealthyIoT 2017, Angers, France, October 24-25, 2017, Proceedings},
        proceedings_a={HEALTHYIOT},
        year={2018},
        month={2},
        keywords={Elderly fall detection Micro electro mechanical system Inertial measurement unit Support vector machine Multi-layer perceptron K-nearest neighbors},
        doi={10.1007/978-3-319-76213-5_5}
    }
    
  • L. Nguyen
    M. Saleh
    R. Bouquin Jeannès
    Year: 2018
    An Efficient Design of a Machine Learning-Based Elderly Fall Detector
    HEALTHYIOT
    Springer
    DOI: 10.1007/978-3-319-76213-5_5
L. Nguyen, M. Saleh, R. Bouquin Jeannès,*
    *Contact email: regine.le-bouquin-jeannes@univ-rennes1.fr

    Abstract

    Elderly fall detection is an important health care application as falls represent the major reason of injuries. An efficient design of a machine learning-based wearable fall detection system is proposed in this paper. The proposed system depends only on a 3-axial accelerometer to capture the elderly motion. As the power consumption is proportional to the sampling frequency, the performance of the proposed fall detector is analyzed as a function of this frequency in order to determine the best trade-off between performance and power consumption. Thanks to efficient extracted features, the proposed system achieves a sensitivity of 99.73% and a specificity of 97.7% using a 40 Hz sampling frequency notably outperforming reference algorithms when tested on a large dataset.