Study on the Prediction and Prevention of Falls Risk in the Elderly based on Neural Network

DOI

10.2991/978-2-38476-346-7_2How to use a DOI?

Keywords

component; Fall risk in the elderly; Neural network; Long short-term memory network; Prediction model

Abstract

Objective: To improve the accuracy of fall risk prediction for the elderly, establish an efficient prediction model based on neural network, and thus reduce the number of fall events, improve the quality of life and safety level of the elderly. Methods: A comprehensive study of the sports, health and environmental data of 500 industry experts over a one-year time span was conducted. A multi-level perception analysis model, convolutional neural network (CNN) and long short-term memory network (LSTM) were used for comparative evaluation. Ten rounds of cross-validation and Adam algorithm were used to improve the model, and its precision, positive prediction accuracy, true positive rate and F1-score were estimated. Results: The long short-term memory network model performed well in many evaluation criteria, with an accuracy of 93%, a positive prediction rate of 91%, a detection rate of 92%, and an F1-score of 91%. Conclusion: The LSTM neural network has excellent performance in the identification of fall signs in the elderly population, and has the potential to be used for early warning and customized prevention plans to reduce the frequency of falls.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Lei Liu
PY  - 2024
DA  - 2024/12/27
TI  - Study on the Prediction and Prevention of Falls Risk in the Elderly based on Neural Network
BT  - Proceeding of the 2024 International Conference on Diversified Education and Social Development (DESD 2024)
PB  - Atlantis Press
SP  - 4
EP  - 11
SN  - 2352-5398
UR  - https://doi.org/10.2991/978-2-38476-346-7_2
DO  - 10.2991/978-2-38476-346-7_2
ID  - Liu2024
ER  -