Dataset Information

Application of Wearable Inertial Sensors and A New Test Battery for Distinguishing Retrospective Fallers from Non-fallers among Community-dwelling Older People.

ABSTRACT: Considering the challenge of population ageing and the substantial health problem among the elderly population from falls, the purpose of this study was to verify whether it is possible to distinguish accurately between older fallers and non-fallers, based on data from wearable inertial sensors collected during a specially designed test battery. A comprehensive but practical test battery using 5 wearable inertial sensors for multifactorial fall risk assessment was designed. This was followed by an experimental study on 196 community-dwelling Korean older women, categorized as fallers (N₁?=?82) and non-fallers (N₂?=?114) based on prior history of falls. Six machine learning models (logistic regression, naïve bayes, decision tree, random forest, boosted tree and support vector machine) were proposed for faller classification. Results indicated that compared with non-fallers, fallers performed significantly worse on the test battery. In addition, the application of sensor data and support vector machine for faller classification achieved an overall accuracy of 89.4% with 92.7% sensitivity and 84.9% specificity. These findings suggest that wearable inertial sensor based systems show promise for elderly fall risk assessment, which could be implemented in clinical practice to identify "at-risk" individuals reliably to promote proactive fall prevention.

SUBMITTER: Qiu H

PROVIDER: S-EPMC6218502 | biostudies-literature | 2018 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Application of Wearable Inertial Sensors and A New Test Battery for Distinguishing Retrospective Fallers from Non-fallers among Community-dwelling Older People.

Qiu Hai H Rehman Rana Zia Ur RZU Yu Xiaoqun X Xiong Shuping S

Scientific reports 20181105 1

Considering the challenge of population ageing and the substantial health problem among the elderly population from falls, the purpose of this study was to verify whether it is possible to distinguish accurately between older fallers and non-fallers, based on data from wearable inertial sensors collected during a specially designed test battery. A comprehensive but practical test battery using 5 wearable inertial sensors for multifactorial fall risk assessment was designed. This was followed by ...[more]

PMID: 30397282

Similar Datasets

Project description:BACKGROUND:Fall rates among adults with multiple sclerosis are consistently greater than 50%, but near-falls (i.e. a trip or stumble) are often undocumented. Furthermore, little is known about the circumstances surrounding fall and near-fall events. The purpose of this study was to examine the similarities and differences among non-fallers, near-fallers and fallers with multiple sclerosis, including the circumstances that surround falls and near-falls. METHODS:In a single visit, 135 multiple sclerosis participants completed the Hopkins Falls Grading Scale, a custom questionnaire investigating circumstances surrounding falls and near-falls, and performed the Timed Up and Go and Timed 25-Foot Walk tests. Mann-Whitney tests were used to examine differences between fallers, near-fallers and non-fallers. Multiple logistic regression with AIC criterion was used to examine associations of circumstances with the odds of falling vs. near-falling. Cumulative odds ordinal logistic regression was used to analyze the association between each of the walking tests and the susceptibility of the individual for falls or near-falls. RESULTS:30% of individuals reported falls, while 44% reported near-falls over a 1-year period. Non-fallers completed the walking tests more quickly than near-fallers (p < 0.0045), and fallers (p < 0.0001); near-fallers and fallers demonstrated similar motor profiles. Individuals were more likely to sustain a fall rather than a near-fall under the following circumstances: transferring outside the home (p = 0.015) and tripping over an obstacle (p = 0.025). Performing 1-second slower on the walking tests increased the odds of a history of a fall by 6-20%. CONCLUSION:Near-falls occur commonly in individuals with MS; near-fallers and fallers reported similar circumstances surrounding fall events and demonstrated similar performance on standard timed walking tests. Clinicians monitoring individuals with MS should consider evaluation of the circumstances surrounding falls in combination with quantitative walking measures to improve determination of fall risk and appropriate rehabilitation interventions.

Project description:Falls in the elderly is a major public health concern due to its high prevalence, serious consequences and heavy burden on the society. Many falls in older people happen within a very short time, which makes it difficult to predict a fall before it occurs and then to provide protection for the person who is falling. The primary objective of this study was to develop deep neural networks for predicting a fall during its initiation and descending but before the body impacts to the ground so that a safety mechanism can be enabled to prevent fall-related injuries. We divided the falling process into three stages (non-fall, pre-impact fall and fall) and developed deep neutral networks to perform three-class classification. Three deep learning models, convolutional neural network (CNN), long short term memory (LSTM), and a novel hybrid model integrating both convolution and long short term memory (ConvLSTM) were proposed and evaluated on a large public dataset of various falls and activities of daily living (ADL) acquired with wearable inertial sensors (accelerometer and gyroscope). Fivefold cross validation results showed that the hybrid ConvLSTM model had mean sensitivities of 93.15, 93.78, and 96.00% for non-fall, pre-impact fall and fall, respectively, which were higher than both LSTM (except the fall class) and CNN models. ConvLSTM model also showed higher specificities for all three classes (96.59, 94.49, and 98.69%) than LSTM and CNN models. In addition, latency test on a microcontroller unit showed that ConvLSTM model had a short latency of 1.06 ms, which was much lower than LSTM model (3.15 ms) and comparable with CNN model (0.77 ms). High prediction accuracy (especially for pre-impact fall) and low latency on the microboard indicated that the proposed hybrid ConvLSTM model outperformed both LSTM and CNN models. These findings suggest that our proposed novel hybrid ConvLSTM model has great potential to be embedded into wearable inertial sensor-based systems to predict pre-impact fall in real-time so that protective devices could be triggered in time to prevent fall-related injuries for older people.

Project description:BackgroundPeople with mobility impairments may have difficulties in everyday life motor activities, and assessing these difficulties is crucial to plan rehabilitation interventions and evaluate their effectiveness. Wearable inertial sensors enable long-term monitoring of motor activities in a patient's habitual environment and complement clinical assessments which are conducted in a standardised environment. The application of wearable sensors requires appropriate data processing algorithms to estimate clinically meaningful outcome measures, and this review will provide an overview of previously published measures, their underlying algorithms, sensor placement, and measurement properties such as validity, reproducibility, and feasibility.MethodsWe will screen the literature for studies which applied inertial sensors to people with mobility impairments in free-living conditions, described the data processing algorithm reproducibly, and calculated everyday life motor activity-related outcome measures. Three databases (MEDLINE, EMBASE, and SCOPUS) will be searched with terms out of four different categories: study population, measurement tool, algorithm, and outcome measure. Abstracts and full texts will be screened independently by the two review authors, and disagreement will be solved by discussion and consensus. Data will be extracted by one of the review authors and verified by the other. It includes the type of outcome measures, the underlying data processing algorithm, the required sensor technology, the corresponding sensor placement, the measurement properties, and the target population. We expect to find a high heterogeneity of outcome measures and will therefore provide a narrative synthesis of the extracted data.DiscussionThis review will facilitate the selection of an appropriate sensor setup for future applications, contain recommendations about the design of data processing algorithms as well as their evaluation procedure, and present a gap for innovative, new algorithms, and devices.Systematic review registrationInternational prospective register of systematic reviews (PROSPERO): CRD42017069865 .

Dataset Information

Application of Wearable Inertial Sensors and A New Test Battery for Distinguishing Retrospective Fallers from Non-fallers among Community-dwelling Older People.

Publications

Application of Wearable Inertial Sensors and A New Test Battery for Distinguishing Retrospective Fallers from Non-fallers among Community-dwelling Older People.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets