BROWSE
Detail
Conference Paper
A Study on CNN-Based Berg Balance Scale Analysis for Elderly Persons
Cited 2 time in 
Download 1 time
Share 
Download 1 time
Share
- Authors
- Yeonsu Lee, Sungjae Yoon, Wonjong Kim
- Issue Date
- 2019-06
- Citation
- International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC) 2019, pp.252-253
- Language
- English
- Type
- Conference Paper
- Project Code
- 19ZT1100, ICT convergence technology supporting project for metropolitan industry, Na Jung-Chan
- Abstract
- In this paper, we propose a method to analyze berg balance scale (BBS) for elderly persons using convolutional neural network (CNN). One of the main problem for elderly people is that their health is easily get deteriorated after a fall. The proposed CNN model predicts the BBS score of the elderly using sensors that can measure the motion pattern while performing activities of BBS. We developed two CNN models - one dimensional and two dimensional - to predict the score of BBS activities. The models predict the score between 0 to 4 with the accuracy of 90%. The proposed CNN models can be used to evaluate fall risk for elderly persons.
- KSP Keywords
- CNN model, Convolution neural network(CNN), Elderly People, Elderly person, Motion Pattern, One-dimensional, berg balance scale, fall risk, scale analysis, two-dimensional(2D)