Deep neural networks for human’s fall-risk prediction using force-plate time series signal

作者：

Highlights：

• Human’s balance ability is predictable using deep neural networks.

• C-LSTM provides a better prediction of human balance patterns.

• Human’s risk of fall can be characterized based on force-plate time series metrics.

• Random sampling and architectural simplicity improve neural network performance.

摘要

•Human’s balance ability is predictable using deep neural networks.•C-LSTM provides a better prediction of human balance patterns.•Human’s risk of fall can be characterized based on force-plate time series metrics.•Random sampling and architectural simplicity improve neural network performance.

论文关键词：Aging,Balance disorder,Balance impairment,CNN,C-LSTM,Deep learning,Fall-risk,Force-plate,LSTM,Neural network,RNN

论文评审过程：Received 10 October 2020, Revised 11 March 2021, Accepted 13 May 2021, Available online 26 May 2021, Version of Record 7 June 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115220