Real-Life/Real-Time Elderly Fall Detection with a Triaxial Accelerometer

Abstract

The consequences of a fall on an elderly person can be reduced if the accident is attended by medical personnel within the first hour. Independent elderly people often stay alone for long periods of time, being in more risk if they suffer a fall. The literature offers several approaches for detecting falls with embedded devices or smartphones using a triaxial accelerometer. Most of these approaches have not been tested with the target population or cannot be feasibly implemented in real-life conditions. In this work, we propose a fall detection methodology based on a non-linear classification feature and a Kalman filter with a periodicity detector to reduce the false positive rate. This methodology requires a sampling rate of only 25 Hz; it does not require large computations or memory and it is robust among devices. We tested our approach with the SisFall dataset achieving 99.4% of accuracy. We then validated it with a new round of simulated activities with young adults and an elderly person. Finally, we give the devices to three elderly persons for full-day validations. They continued with their normal life and the devices behaved as expected.

Fall Detection Using Multiple Bioradars and Convolutional Neural Networks

“…The achieved results should be accepted with caution because the experimental dataset used for the classifier training is relatively small and acquired only for young volunteers. Nevertheless, in [42], it is shown that datasets for young people can be used instead of the one for elderly people while training a fall classifier.…”

A User-Adaptive Algorithm for Activity Recognition Based on K-Means Clustering, Local Outlier Factor, and Multivariate Gaussian Distribution

“…Three thresholds for resultant angular velocity, resultant angular acceleration, and resultant angle change were set to distinguish falls from activities of daily living. Sucerquia et al [24] presented a novel fall detection approach based on a Kalman filter and a non-linear classification feature, using data from a tri-axial accelerometer. Their methodology required a low sampling frequency of only 25 Hz.…”

“…Shi et al [27] uses J48 decision tree, which is an efficient algorithm derived from C4.5 decision tree [28], to detect falls. The aforementioned methods [22,23,24,25,26,27] do not consider the effect of personalization on fall detection. Medrano et al [8] evaluated four algorithms (nearest neighbor (NN), local outlier factor (LOC), one-class support vector machine (One-Class SVM), and SVM) after personalization used as fall detectors to boost their performance when compared to their non-personalized versions.…”

The Use of Wearable Sensors in Human Movement Analysis in Non-Swimming Aquatic Activities: A Systematic Review

“…The use of smart technology applied to the human movement/exercise analysis has been a must for researchers, clinicians, practitioners, and patients. Inertial measurement unit technology is a less time consuming, noninvasive, and practical alternative to the video-based methods used by researchers to analyze human motion, allowing to detect with a higher precision several parameters during exercise [1,2]. Technology advanced in a way that due to the sensor' compactness (small size) multiple applications may be used in human locomotion, including exercise/sports movements [3,4].…”