Wearable Embedded Intelligence for Detection of Falls Independently of on-Body Location

Abstract

Falls are one of the most common problems in the elderly population. Therefore, each year more solutions for automatic fall detection are emerging. This paper proposes a single accelerometer algorithm for wearable devices that works for three different body locations: chest, waist and pocket, without a calibration step being required. This algorithm is able to be fully executed on a wearable device and no external devices are necessary for data processing. Additionally, a study of the accelerometer sampling rate, that allows the algorithm to achieve a better performance, was performed. The algorithm was validated with a continuous dataset with daily living activities and 272 simulated falls. Considering the trade-off between sensitivity and the number of false alarms the most suitable sampling rate found was 50 Hz. The proposed algorithm was able to achieve a trade-off of no false alarms and 89.5% of fall detection rate when wearing the sensor on the user’s waist with a medium sensitivity level of the algorithm. In conclusion, this paper presents a reliable solution for automatic fall detection that can be adapted to different usages and conditions, since it can be used in different body locations and its sensitivity can be adapted to different subjects according to their physical activity level.

Automated Development of Custom Fall Detectors: Position, Model and Rate Impact in Performance

“…Most previous works have acquired data from simulated falls and ADLs. Besides acquiring scripted samples in laboratory conditions, other studies have focused on acquiring and evaluating the trained models in free-living scenarios, from a continuous usage of the wearable devices, like Alves et al [5]. Nevertheless, fall detection is usually an unbalanced problem, with a higher percentage of non-falls compared to falls in most datasets reported in prior works.…”

Fusion of Video and Inertial Sensing for Deep Learning–Based Human Action Recognition

“…Vision and inertial sensing modalities have been used individually to achieve human action recognition, e.g., [13,14,15,16,17,18,19,20,21,22,23]. Furthermore, the use of deep learning models or deep neural networks have proven to be more effective than conventional approaches for human action recognition.…”

“…The main advantage of these sensors include their wearability—Thus, they are not limited to a specific field of view. Often, 3-axis acceleration signals from their accelerometers and 3-axis angular velocity signals from their gyroscopes are used for conducting human action recognition, e.g., [17,18,19,20,21,22,23]. These sensors have also limitations in terms of not capturing a complete representation of actions.…”

A Technological-Based Platform for Risk Assessment, Detection, and Prevention of Falls Among Home-Dwelling Older Adults: Protocol for a Quasi-Experimental Study

“…In addition to the variables retrieved by the clinical application (eg, personal profile, medical conditions, medication, answers to the questionnaires, and scores of the 3 functional tests, the home application will allow the measurement of the range of motion along with the number of repetitions and durations of ascending and descending movements for the 8 exercises of the OEP, namely, knee flexion, knee extension, hip abduction, knee bending, toe raises, calf raises, sit-to-stand, and one-leg standing exercises. Previous studies have set a background for the technological solutions used in this study [45,46]. This study has some limitations, namely, the use of a nonrandom sample and the absence of a control group.…”