EUVIMEDEUVIMED entdecken
Open Access

An Effective Deep Learning Framework for Fall Detection: Model Development and Study Design

Jinxi Zhang iD, Zhen Li iD, Yu Liu iD, Jian Li iD, Hualong Qiu iD, Mohan Li iD, Guohui Hou iD, Zhixiong Zhou iD

Journal of Medical Internet Research · 05.08.2024 · 26 · e56750

DOI: 10.2196/56750 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Medical Internet Research, online erschienen am 05.08.2024. Verfasst von Jinxi Zhang und 7 weiteren Autor:innen. Beteiligte Einrichtung: Beijing Kupei Sports Culture Corporation Limited, Beijing, China. Hauptthemen laut MEDLINE: Accidental Falls, Deep Learning. Die Arbeit wurde bisher 51-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Fall detection is of great significance in safeguarding human health. By monitoring the motion data, a fall detection system (FDS) can detect a fall accident. Recently, wearable sensors–based FDSs have become the mainstream of research, which can be categorized into threshold-based FDSs using experience, machine learning–based FDSs using manual feature extraction, and deep learning (DL)–based FDSs using automatic feature extraction. However, most FDSs focus on the global information of sensor data, neglecting the fact that different segments of the data contribute variably to fall detection. This shortcoming makes it challenging for FDSs to accurately distinguish between similar human motion patterns of actual falls and fall-like actions, leading to a decrease in detection accuracy.

Objective

This study aims to develop and validate a DL framework to accurately detect falls using acceleration and gyroscope data from wearable sensors. We aim to explore the essential contributing features extracted from sensor data to distinguish falls from activities of daily life. The significance of this study lies in reforming the FDS by designing a weighted feature representation using DL methods to effectively differentiate between fall events and fall-like activities.

Methods

Based on the 3-axis acceleration and gyroscope data, we proposed a new DL architecture, the dual-stream convolutional neural network self-attention (DSCS) model. Unlike previous studies, the used architecture can extract global feature information from acceleration and gyroscope data. Additionally, we incorporated a self-attention module to assign different weights to the original feature vector, enabling the model to learn the contribution effect of the sensor data and enhance classification accuracy. The proposed model was trained and tested on 2 public data sets: SisFall and MobiFall. In addition, 10 participants were recruited to carry out practical validation of the DSCS model. A total of 1700 trials were performed to test the generalization ability of the model.

Results

The fall detection accuracy of the DSCS model was 99.32% (recall=99.15%; precision=98.58%) and 99.65% (recall=100%; precision=98.39%) on the test sets of SisFall and MobiFall, respectively. In the ablation experiment, we compared the DSCS model with state-of-the-art machine learning and DL models. On the SisFall data set, the DSCS model achieved the second-best accuracy; on the MobiFall data set, the DSCS model achieved the best accuracy, recall, and precision. In practical validation, the accuracy of the DSCS model was 96.41% (recall=95.12%; specificity=97.55%).

Conclusions

This study demonstrates that the DSCS model can significantly improve the accuracy of fall detection on 2 publicly available data sets and performs robustly in practical validation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Accidental FallsDeep LearningHumansWearable Electronic DevicesNeural Networks, ComputerMale

Stichwörter der Autor:innen

MobiFallSisfallaccelerometerdeep learningfall detectiongyroscopehuman healthself-attentionwearable sensors

Autor:innen und Institutionen

  • Beijing Kupei Sports Culture Corporation Limited, Beijing, China.
  • Institute of Artificial Intelligence in Sports, Capital University of Physical Education and Sports, Beijing, China.
  • School of Physical Education and Sport Science, Fujian Normal University, Fuzhou, China.
  • Institute for Sport Performance and Health Promotion, Capital University of Physical Education and Sports, Beijing, China.
  • Bioelectronics Center of YZW, Shanghai, China.
  • Walt Technology Group Co, Ltd, Jiaxing, China.

ORCID-Profile: Jinxi Zhang, Zhen Li, Yu Liu, Jian Li, Hualong Qiu, Mohan Li, Guohui Hou, Zhixiong Zhou

Zitieren

APA

Zhang, J., Li, Z., Liu, Y., Li, J., Qiu, H., Li, M., Hou, G., & Zhou, Z. (2024). An Effective Deep Learning Framework for Fall Detection: Model Development and Study Design. Journal of Medical Internet Research, 26, e56750. https://doi.org/10.2196/56750

Vancouver

Zhang J, Li Z, Liu Y, Li J, Qiu H, Li M, et al. An Effective Deep Learning Framework for Fall Detection: Model Development and Study Design. Journal of Medical Internet Research. 2024;26:e56750. doi:10.2196/56750

BibTeX
@article{Zhang2024,
  author = {Zhang, Jinxi and Li, Zhen and Liu, Yu and Li, Jian and Qiu, Hualong and Li, Mohan and Hou, Guohui and Zhou, Zhixiong},
  title = {An Effective Deep Learning Framework for Fall Detection: Model Development and Study Design},
  journal = {Journal of Medical Internet Research},
  year = {2024},
  volume = {26},
  pages = {e56750},
  doi = {10.2196/56750}
}
Kostenlos starten →