A Novel Monitoring System for Fall Detection in Older People
Open accessThe DOI matches its Crossref record; 34 of the 34 references it cites were found in a registry or an index.
Abstract
Each year, more than 30% of people over 65 years-old suffer some fall. Unfortunately, this can generate physical and psychological damage, especially if they live alone and they are unable to get help. In this field, several studies have been performed aiming to alert potential falls of the older people by using different types of sensors and algorithms. In this paper, we present a novel non-invasive monitoring system for fall detection in older people who live alone. Our proposal is using very-low-resolution thermal sensors for classifying a fall and then alerting to the care staff. Also, we analyze the performance of three recurrent neural networks for fall detections: long short-term memory (LSTM), gated recurrent unit, and Bi-LSTM. As many learning algorithms, we have performed a training phase using different test subjects. After several tests, we can observe that the Bi-LSTM approach overcome the others techniques reaching a 93% of accuracy in fall detection. We believe that the bidirectional way of the Bi-LSTM algorithm gives excellent results because the use of their data is influenced by prior and new information, which compares to LSTM and GRU. Information obtained using this system did not compromise the user's privacy, which constitutes an additional advantage of this alternative.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · cited 1×
“This recurrent neural network was introduced and designed byHochreiter and Schmedhuber (1997) [33] to handle temporal data sequences and also to be able to confront the exploding and vanishing gradient problems, which is a recurring problem with traditional RNNs.”
(2014) · Cited by 33,419
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S., & Szegedy, C. (2015) · arXiv · doi:10.48550/arxiv.1502.03167 · Cited by 27,818 · cited 1×
“process [37], making the model conform to the specific distribution of each mini batch.”
MethodLearning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation
Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1179 · Cited by 25,545 · cited 2×
“These are LSTM, GRU, and Bi-LSTM; we have used these RNNs because they have been highly used for classification problems [24]–[26].”
MethodClegg, A., Young, J., Iliffe, S., Rikkert, M. O., & Rockwood, K. (2013) · The Lancet · doi:10.1016/s0140-6736(12)62167-9 · Cited by 9,503
Gate-variants of Gated Recurrent Unit (GRU) neural networks
Dey, R., & Salem, F. M. (2017) · 2017 IEEE 60th International Midwest Symposium on Circuits and Systems (MWSCAS) · doi:10.1109/mwscas.2017.8053243 · Cited by 2,186
A Deep Learning Approach for Intrusion Detection Using Recurrent Neural Networks
Yin, C., Zhu, Y., Fei, J., & He, X. (2017) · IEEE Access · doi:10.1109/access.2017.2762418 · Cited by 2,100 · cited 2×
“Table 4 summarizes the results obtained with each of the implemented models, trained with the published datasets in [30]: From the architectures tested, satisfactory results were obtained for each of the models.”
BackgroundMethodRisk factors for falls among older adults: A review of the literature
Ambrose, A. F., Paul, G., & Hausdorff, J. M. (2013) · Maturitas · doi:10.1016/j.maturitas.2013.02.009 · Cited by 1,888
Preventing Falls in Elderly Persons
Tinetti, M. E. (2003) · New England Journal of Medicine · doi:10.1056/nejmcp020719 · Cited by 1,660
LSTM Fully Convolutional Networks for Time Series Classification
Karim, F., Majumdar, S., Darabi, H., & Chen, S. (2018) · IEEE Access · doi:10.1109/access.2017.2779939 · Cited by 1,465 · cited 2×
“The model of this neural network can be described by the following equations: Where While LSTMs possess the ability to learn temporal dependencies in sequences, they have difficulty with long term dependencies in long sequences [24].”
MethodThe coming acceleration of global population ageing
Lutz, W., Sanderson, W., & Scherbov, S. (2008) · Nature · doi:10.1038/nature06516 · Cited by 1,437 · cited 1×
“INTRODUCTION The effects of fertility decline, along with increased life expectancy, portend the acceleration of global population aging [1]–[3].”
BackgroundBidirectional LSTM Networks for Improved Phoneme Classification and Recognition
Graves, A., Fernández, S., & Schmidhuber, J. (2005) · Lecture Notes in Computer Science · doi:10.1007/11550907_126 · Cited by 925 · cited 1×
“It is equipped with two parallel layers called backward and forward, which can pass information in the same manner that LSTM does, but in both directions [25].”
BackgroundImplications of population ageing for economic growth
Bloom, D. E., Canning, D., & Fink, G. (2010) · Oxford Review of Economic Policy · doi:10.1093/oxrep/grq038 · Cited by 870
Global aging : the challenge of success
Kinsella, K., & Phillips, D. R. (2005) · Digital Commons - Lingnan (Lingnan University) · Cited by 786
Preventing Falls and Fall-Related Injuries in Hospitals
Oliver, D., Healey, F., & Haines, T. P. (2010) · Clinics in Geriatric Medicine · doi:10.1016/j.cger.2010.06.005 · Cited by 480 · cited 1×
“According to [12], there is an estimation of 3 to 5 falls per 1,000 days of stay in a hospital.”
BackgroundA smart sensor to detect the falls of the elderly
Sixsmith, A., & Johnson, N. (2004) · IEEE Pervasive Computing · doi:10.1109/mprv.2004.1316817 · Cited by 399 · cited 5×
“Sixsmith and Johnson [22] performed one of the first studies using low-resolution infrared sensor arrays (16× 16 pixels).”
BackgroundMethodInfluentialApproaches and principles of fall detection for elderly and patient
Yu, X. (2008) · HealthCom 2008 - 10th International Conference on e-health Networking, Applications and Services · doi:10.1109/health.2008.4600107 · Cited by 328 · cited 1×
“In terms of sensor technology developed up to date, each proposed system can be classified into: a wearable-based system, a camera-based system, and an ambience device [20].”
BackgroundDavis, J. C., Robertson, M. C., Ashe, M. C., Liu-Ambrose, T., Khan, K. M., & Marra, C. A. (2010) · Osteoporosis International · doi:10.1007/s00198-009-1162-0 · Cited by 311
Fast image scanning with deep max-pooling convolutional neural networks
Giusti, A., Cireşan, D. C., Masci, J., Gambardella, L. M., & Schmidhuber, J. (2013) · 2013 IEEE International Conference on Image Processing · doi:10.1109/icip.2013.6738831 · Cited by 294 · cited 1×
“This layer is formed by a 1D MaxPooling with a 2-pool size and has the ability to reduce the number of characteristics encountered, leaving only the ones of interest for the network and therefore obtaining a better capacity for generalization [32].”
BackgroundFinlayson, M. L., & Peterson, E. W. (2010) · Physical Medicine and Rehabilitation Clinics of North America · doi:10.1016/j.pmr.2009.12.003 · Cited by 162 · cited 1×
“A significant proportion of these persons lives alone, which makes obtaining help in the case of a fall much more difficult [8]–[11].”
BackgroundVision-based fall detection system for improving safety of elderly people
Harrou, F., Zerrouki, N., Sun, Y., & Houacine, A. (2017) · IEEE Instrumentation & Measurement Magazine · doi:10.1109/mim.2017.8121952 · Cited by 130 · cited 2×
“For example, [19], formulates the use of video cameras to detect falls through the implementation of Exponentially Weighted Moving Average based on Support Vector Machines (MEWMA-SVM), which achieved high precision in fall detection (97.”
BackgroundMethodGeriatric Syndromes and Geriatric Assessment for the Generalist
Carlson, C. M., Merel, S. E., & Yukawa, M. (2015) · Medical Clinics of North America · doi:10.1016/j.mcna.2014.11.003 · Cited by 117
Min, W., Cui, H., Rao, H., Li, Z., & Yao, L. (2018) · IEEE Access · doi:10.1109/access.2018.2795239 · Cited by 109 · cited 1×
“Recently, numerous studies in the area of fall detection have arisen in an attempt to solve this problem [17]–[19].”
BackgroundSampalis, J. S., Nathanson, R., Vaillancourt, J., Nikolis, A., Liberman, M., Angelopoulos, J., Krassakopoulos, N., Longo, N., & Psaradellis, E. (2009) · Annals of Surgery · doi:10.1097/sla.0b013e31819a8b4f · Cited by 87 · cited 1×
“Moreover, the mortality rate of older people who have suffered some sort of fall is significantly higher than those who have not suffered any falls, with an odds ratio (OR) of 5.11 (CI 95% 1.84 - 14.17, p = 0.002) [14]–[16].”
BackgroundInfluentialHsieh, Y., & Jeng, Y.-L. (2017) · IEEE Access · doi:10.1109/access.2017.2771389 · Cited by 87
Falls in the elderly of the Family Health Program
Coimbra, A. M. V., Ricci, N. A., Coimbra, I. B., & Costallat, L. T. L. (2010) · Archives of Gerontology and Geriatrics · doi:10.1016/j.archger.2010.01.010 · Cited by 77 · cited 2×
“Falls are especially relevant to patients and health systems because approximately one-third of adults older than 65 that live in a community suffer a fall each year [7]–[11].”
BackgroundLow-Power Fall Detector Using Triaxial Accelerometry and Barometric Pressure Sensing
Wang, C., Lü, W., Narayanan, M. R., Chang, D. C. W., Lord, S. R., Redmond, S. J., & Lovell, N. H. (2016) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2016.2587761 · Cited by 75 · cited 1×
“[21] propose the use of a wearable device with an estimated battery life of 664.”
BackgroundLeveraging Biomedical Resources in Bi-LSTM for Drug-Drug Interaction Extraction
Xu, B., Shi, X., Zhao, Z., & Zheng, W. (2018) · IEEE Access · doi:10.1109/access.2018.2845840 · Cited by 65 · cited 1×
“LSTM network generates two states at every single time-step: a cell state that is transferred into the next time-step and a hidden state that is the output vector of the time-step [34].”
BackgroundAcoustic cues from the floor: A new approach for fall classification
Principi, E., Droghini, D., Squartini, S., Olivetti, P., & Piazza, F. (2016) · Expert Systems with Applications · doi:10.1016/j.eswa.2016.04.007 · Cited by 50
Robust unobtrusive fall detection using infrared array sensors
Fan, X., Zhang, H., Leung, C., & Shen, Z. (2017) · IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI) · doi:10.1109/mfi.2017.8170428 · Cited by 50 · cited 2×
“The use of lowresolution thermal sensors is a viable alternative for this purpose, given that they are cost-effective, non-intrusive, and can yield distinct types of information such as position, velocity, acceleration, and human body temperature in a controlled environment [22], [23], [27], [28].”
BackgroundNavaratnarajah, A., & Jackson, S. H. (2012) · Medicine · doi:10.1016/j.mpmed.2012.10.009 · Cited by 45
A fall detection system using low resolution infrared array sensor
Mashiyama, S., Hong, J., & Ohtsuki, T. O. (2014) · doi:10.1109/pimrc.2014.7136520 · Cited by 41 · cited 1×
“The use of lowresolution thermal sensors is a viable alternative for this purpose, given that they are cost-effective, non-intrusive, and can yield distinct types of information such as position, velocity, acceleration, and human body temperature in a controlled environment [22], [23], [27], [28].”
BackgroundAn Ensemble of Convolutional Neural Networks for Image Classification Based on LSTM
Chen, J., Wang, Y., Wu, Y., & Cai, C. (2017) · International Conference on Green Informatics (ICGI) · doi:10.1109/icgi.2017.36 · Cited by 28 · cited 1×
“The convolutional layer is implemented with the ReLU activation function due to its characteristics that allow it to reduce the effects of vanishing gradient, achieving faster learning [31].”
MethodA falling detection system with plural thermal array sensors
Taniguchi, Y., Nakajima, H., Tsuchiya, N., Tanaka, J., Aita, F., & Hata, Y. (2014) · Joint International Conference on Soft Computing and Intelligent Systems and International Symposium on Advanced Intelligent Systems (SCIS&ISIS) · doi:10.1109/scis-isis.2014.7044834 · Cited by 22 · cited 3×
“The use of lowresolution thermal sensors is a viable alternative for this purpose, given that they are cost-effective, non-intrusive, and can yield distinct types of information such as position, velocity, acceleration, and human body temperature in a controlled environment [22], [23], [27], [28].”
BackgroundMethod
Research works printed but not linked: 3
- 2018, NDN Github Repository
- 2018, Falls
- 2015, World Population Ageing 2015
Cited by
Virginia Anikwe, C., Friday Nweke, H., Chukwu Ikegwu, A., Adolphus Egwuonwu, C., Uchenna Onu, F., Rita Alo, U., & Wah Teh, Y. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.117362 · Cited by 233
A Survey on Recent Advances in Wearable Fall Detection Systems
Ramachandran, A., & Karuppiah, A. (2020) · BioMed Research International · doi:10.1155/2020/2167160 · Cited by 173
“visual deficit [9]; mobility limitation [8]; cognitive impairment [5]; impaired functional status [4];”
BackgroundDeep Learning Based Systems Developed for Fall Detection: A Review
Islam, M. M., Tayan, O., Islam, M. R., Islam, M. S., Nooruddin, S., Nomani Kabir, M., & Islam, M. R. (2020) · IEEE Access · doi:10.1109/access.2020.3021943 · Cited by 159
A Framework for Fall Detection Based on OpenPose Skeleton and LSTM/GRU Models
Lin, C.-B., Dong, Z., Kuan, W.-K., & Huang, Y. (2020) · Applied Sciences · doi:10.3390/app11010329 · Cited by 154
Badidi, E., Moumane, K., & Ghazi, F. E. (2023) · IEEE Access · doi:10.1109/access.2023.3300658 · Cited by 150
“In [145], the authors implemented a non-invasive system based on RNNs and video streams to monitor fall detection in elderly living alone.”
Background
Metadata
- Authors
- Taramasco, C., Rodenas, T., Martinez, F., Fuentes, P., Munoz, R., Olivares, R., De Albuquerque, V. H. C., & Demongeot, J.
- Affiliations
- University of Valparaíso (CL)
- Universidade de Fortaleza (BR)
- Venue
- IEEE Access
- Published
- 2018
- Volume
- 6
- Pages
- 43563-43574
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W2889528477
- Cited by
- 113
- References
- 34
- Citations by year
- 2026: 4 · 2025: 10 · 2024: 15 · 2023: 21 · 2022: 17 · 2021: 17 · 2020: 18 · 2019: 10 · 2018: 1
- FWCI
- 5.76
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- IEEE Access (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Repositorio Institucional Académico , Sistema de Bibliotecas Universidad Andrés Bello (UNAB) (Universidad Andrés Bello (UNAB)) (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Health Sciences › Health Professions › Physical Therapy, Sports Therapy and Rehabilitation › Balance, Gait, and Falls Prevention
- Physical Sciences › Engineering › Biomedical Engineering › Non-Invasive Vital Sign Monitoring
- Keywords
- fall detection, older adults, temperature sensors, non-invasive monitoring, bidirectional LSTM, recurrent neural networks, gated recurrent unit, long short-term memory, elderly living alone, user privacy
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.