Cargando…

A Study on the Application of Convolutional Neural Networks to Fall Detection Evaluated with Multiple Public Datasets

Due to the repercussion of falls on both the health and self-sufficiency of older people and on the financial sustainability of healthcare systems, the study of wearable fall detection systems (FDSs) has gained much attention during the last years. The core of a FDS is the algorithm that discriminat...

Descripción completa

Detalles Bibliográficos
Autores principales: Casilari, Eduardo, Lora-Rivera, Raúl, García-Lagos, Francisco
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7085732/
https://www.ncbi.nlm.nih.gov/pubmed/32155936
http://dx.doi.org/10.3390/s20051466
_version_ 1783509000362393600
author Casilari, Eduardo
Lora-Rivera, Raúl
García-Lagos, Francisco
author_facet Casilari, Eduardo
Lora-Rivera, Raúl
García-Lagos, Francisco
author_sort Casilari, Eduardo
collection PubMed
description Due to the repercussion of falls on both the health and self-sufficiency of older people and on the financial sustainability of healthcare systems, the study of wearable fall detection systems (FDSs) has gained much attention during the last years. The core of a FDS is the algorithm that discriminates falls from conventional Activities of Daily Life (ADLs). This work presents and evaluates a convolutional deep neural network when it is applied to identify fall patterns based on the measurements collected by a transportable tri-axial accelerometer. In contrast with most works in the related literature, the evaluation is performed against a wide set of public data repositories containing the traces obtained from diverse groups of volunteers during the execution of ADLs and mimicked falls. Although the method can yield very good results when it is hyper-parameterized for a certain dataset, the global evaluation with the other repositories highlights the difficulty of extrapolating to other testbeds the network architecture that was configured and optimized for a particular dataset.
format Online
Article
Text
id pubmed-7085732
institution National Center for Biotechnology Information
language English
publishDate 2020
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-70857322020-03-25 A Study on the Application of Convolutional Neural Networks to Fall Detection Evaluated with Multiple Public Datasets Casilari, Eduardo Lora-Rivera, Raúl García-Lagos, Francisco Sensors (Basel) Article Due to the repercussion of falls on both the health and self-sufficiency of older people and on the financial sustainability of healthcare systems, the study of wearable fall detection systems (FDSs) has gained much attention during the last years. The core of a FDS is the algorithm that discriminates falls from conventional Activities of Daily Life (ADLs). This work presents and evaluates a convolutional deep neural network when it is applied to identify fall patterns based on the measurements collected by a transportable tri-axial accelerometer. In contrast with most works in the related literature, the evaluation is performed against a wide set of public data repositories containing the traces obtained from diverse groups of volunteers during the execution of ADLs and mimicked falls. Although the method can yield very good results when it is hyper-parameterized for a certain dataset, the global evaluation with the other repositories highlights the difficulty of extrapolating to other testbeds the network architecture that was configured and optimized for a particular dataset. MDPI 2020-03-06 /pmc/articles/PMC7085732/ /pubmed/32155936 http://dx.doi.org/10.3390/s20051466 Text en © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Casilari, Eduardo
Lora-Rivera, Raúl
García-Lagos, Francisco
A Study on the Application of Convolutional Neural Networks to Fall Detection Evaluated with Multiple Public Datasets
title A Study on the Application of Convolutional Neural Networks to Fall Detection Evaluated with Multiple Public Datasets
title_full A Study on the Application of Convolutional Neural Networks to Fall Detection Evaluated with Multiple Public Datasets
title_fullStr A Study on the Application of Convolutional Neural Networks to Fall Detection Evaluated with Multiple Public Datasets
title_full_unstemmed A Study on the Application of Convolutional Neural Networks to Fall Detection Evaluated with Multiple Public Datasets
title_short A Study on the Application of Convolutional Neural Networks to Fall Detection Evaluated with Multiple Public Datasets
title_sort study on the application of convolutional neural networks to fall detection evaluated with multiple public datasets
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7085732/
https://www.ncbi.nlm.nih.gov/pubmed/32155936
http://dx.doi.org/10.3390/s20051466
work_keys_str_mv AT casilarieduardo astudyontheapplicationofconvolutionalneuralnetworkstofalldetectionevaluatedwithmultiplepublicdatasets
AT lorariveraraul astudyontheapplicationofconvolutionalneuralnetworkstofalldetectionevaluatedwithmultiplepublicdatasets
AT garcialagosfrancisco astudyontheapplicationofconvolutionalneuralnetworkstofalldetectionevaluatedwithmultiplepublicdatasets
AT casilarieduardo studyontheapplicationofconvolutionalneuralnetworkstofalldetectionevaluatedwithmultiplepublicdatasets
AT lorariveraraul studyontheapplicationofconvolutionalneuralnetworkstofalldetectionevaluatedwithmultiplepublicdatasets
AT garcialagosfrancisco studyontheapplicationofconvolutionalneuralnetworkstofalldetectionevaluatedwithmultiplepublicdatasets