Fall detection algorithm for the elderly using acceleration sensors on the shoes

Fatigue Effect on Minimal Toe Clearance and Toe Activity during Walking

This study investigates the effects of fatigue on the process of walking in young adults using the developed clog-integrated sensor system. The developed sensor can simultaneously measure the forefoot activity (FA) and minimum toe clearance (MTC). The FA was evaluated through the change in the contact area captured by a camera using a method based on a light conductive plate. The MTC was derived from the distance between the bottom surface of the clog and ground obtained using a time of flight (TOF) sensor, and the clog posture was obtained using an acceleration sensor. The induced fatigue was achieved by walking on a treadmill at the fastest walking speed. We evaluated the FA and MTC before and after fatigue in both feet for 14 participants. The effects of fatigue manifested in either the FA or MTC of either foot when the results were evaluated by considering the participants individually, although individual variances in the effects of fatigue were observed. In the dominant foot, a significant increase in either the FA or MTC was observed in 13 of the 14 participants. The mean MTC in the dominant foot increased significantly (p = 0.038) when the results were evaluated by considering the participants as a group.

XAI-Fall: Explainable AI for Fall Detection on Wearable Devices Using Sequence Models and XAI Techniques

A fall detection system is vital for the safety of older people, as it contacts emergency services when it detects a person has fallen. There have been various approaches to detect falls, such as using a single tri-axial accelerometer to detect falls or fixing sensors on the walls of a room to detect falls in a particular area. These approaches have two major drawbacks: either (i) they use a single sensor, which is insufficient to detect falls, or (ii) they are attached to a wall that does not detect a person falling outside its region. Hence, to provide a robust method for detecting falls, the proposed approach uses three different sensors for fall detection, which are placed at five different locations on the subject’s body to gather the data used for training purposes. The UMAFall dataset is used to attain sensor readings to train the models for fall detection. Five models are trained corresponding to the five sensors models, and a majority voting classifier is used to determine the output. Accuracy of 93.5%, 93.5%, 97.2%, 94.6%, and 93.1% is achieved on each of the five sensors models, and 92.54% is the overall accuracy achieved by the majority voting classifier. The XAI technique called LIME is incorporated into the system in order to explain the model’s outputs and improve the model’s interpretability.

Deep Learning for Walking Behaviour Detection in Elderly People Using Smart Footwear

The increase in the proportion of elderly in Europe brings with it certain challenges that society needs to address, such as custodial care. We propose a scalable, easily modulated and live assistive technology system, based on a comfortable smart footwear capable of detecting walking behaviour, in order to prevent possible health problems in the elderly, facilitating their urban life as independently and safety as possible. This brings with it the challenge of handling the large amounts of data generated, transmitting and pre-processing that information and analysing it with the aim of obtaining useful information in real/near-real time. This is the basis of information theory. This work presents a complete system aiming at elderly people that can detect different user behaviours/events (sitting, standing without imbalance, standing with imbalance, walking, running, tripping) through information acquired from 20 types of sensor measurements (16 piezoelectric pressure sensors, one accelerometer returning reading for the 3 axis and one temperature sensor) and warn the relatives about possible risks in near-real time. For the detection of these events, a hierarchical structure of cascading binary models is designed and applied using artificial neural network (ANN) algorithms and deep learning techniques. The best models are achieved with convolutional layered ANN and multilayer perceptrons. The overall event detection performance achieves an average accuracy and area under the ROC curve of 0.84 and 0.96, respectively.

WeMos IoT Controller-Based Low-Cost Fall Detection System for Elderly Users

Fall detection systems for the elderly are very important to protect this type of users. The early detection of the fall of the elderly has a major impact on saving their lives and avoiding the deterioration of the negative medical effects resulting from the effect of the patient falling on a hard surface. One of the constraints in fall detection systems are false-negative errors (no fall detection) or false-positive errors (sending a false warning without real fall accident). These errors have to be reduced significantly. In this paper, an innovative method to reduce fall detection system errors is proposed. The system consists of two orientation detection sensors to track the body orientation instead of using a single sensor in the previous systems which enhances the system accuracy and reduces the false-negative and false-positive errors. The system uses a small size IoT-based controller to process the sensor's information and make the alarm decision based on specific thresholds. The output alarm of the system includes an email sent to the caregivers via the embedded Wi-Fi ESP8266 module as well as an SMS message to the caregivers’ phones via GSM modules to ensure that the alarm message arrives in the absence of internet coverage for the patient or the caregiver. The system is powered by a small lithium-Ion battery. All sensors and modules of the system are combined in a small rubber box that can be fixed in a waist belt or the chest rejoin of the user body. Several tests have been made in different procedures. The tests revealed that the new approach improves the accuracy of the system and reduces the possibility of triggering wrong alarms.

E-Textiles for Healthy Ageing

The ageing population has grown quickly in the last half century with increased longevity and declining birth rate. This presents challenges to health services and the wider society. This review paper considers different aspects (e.g., physical, mental, and social well-being) of healthy ageing and how health devices can help people to monitor health conditions, treat diseases and promote social interactions. Existing technologies for addressing non-physical (e.g., Alzheimer's, loneliness) and physical (e.g., stroke, bedsores, and fall) related challenges are presented together with the drivers and constraints of using e-textiles for these applications. E-textiles provide a platform that enables unobtrusive and ubiquitous deployment of sensors and actuators for healthy ageing applications. However, constraints remain on battery, integration, data accuracy, manufacturing, durability, ethics/privacy issues, and regulations. These challenges can only effectively be met by interdisciplinary teams sharing expertise and methods, and involving end users and other key stakeholders at an early stage in the research.

Manual physical balance assistance of therapists during gait training of stroke survivors: characteristics and predicting the timing

Background

During gait training, physical therapists continuously supervise stroke survivors and provide physical support to their pelvis when they judge that the patient is unable to keep his balance. This paper is the first in providing quantitative data about the corrective forces that therapists use during gait training. It is assumed that changes in the acceleration of a patient’s COM are a good predictor for therapeutic balance assistance during the training sessions Therefore, this paper provides a method that predicts the timing of therapeutic balance assistance, based on acceleration data of the sacrum.

Methods

Eight sub-acute stroke survivors and seven therapists were included in this study. Patients were asked to perform straight line walking as well as slalom walking in a conventional training setting. Acceleration of the sacrum was captured by an Inertial Magnetic Measurement Unit. Balance-assisting corrective forces applied by the therapist were collected from two force sensors positioned on both sides of the patient’s hips. Measures to characterize the therapeutic balance assistance were the amount of force, duration, impulse and the anatomical plane in which the assistance took place. Based on the acceleration data of the sacrum, an algorithm was developed to predict therapeutic balance assistance. To validate the developed algorithm, the predicted events of balance assistance by the algorithm were compared with the actual provided therapeutic assistance.

Results

The algorithm was able to predict the actual therapeutic assistance with a Positive Predictive Value of 87% and a True Positive Rate of 81%. Assistance mainly took place over the medio-lateral axis and corrective forces of about 2% of the patient’s body weight (15.9 N (11), median (IQR)) were provided by therapists in this plane. Median duration of balance assistance was 1.1 s (0.6) (median (IQR)) and median impulse was 9.4Ns (8.2) (median (IQR)). Although therapists were specifically instructed to aim for the force sensors on the iliac crest, a different contact location was reported in 22% of the corrections.

Conclusions

This paper presents insights into the behavior of therapists regarding their manual physical assistance during gait training. A quantitative dataset was presented, representing therapeutic balance-assisting force characteristics. Furthermore, an algorithm was developed that predicts events at which therapeutic balance assistance was provided. Prediction scores remain high when different therapists and patients were analyzed with the same algorithm settings. Both the quantitative dataset and the developed algorithm can serve as technical input in the development of (robot-controlled) balance supportive devices.

Body Acceleration as Indicator for Walking Economy in an Ageing Population

BACKGROUND

In adults, walking economy declines with increasing age and negatively influences walking speed. This study aims at detecting determinants of walking economy from body acceleration during walking in an ageing population.

METHODS

35 healthy elderly (18 males, age 51 to 83 y, BMI 25.5±2.4 kg/m2) walked on a treadmill. Energy expenditure was measured with indirect calorimetry while body acceleration was sampled at 60Hz with a tri-axial accelerometer (GT3X+, ActiGraph), positioned on the lower back. Walking economy was measured as lowest energy needed to displace one kilogram of body mass for one meter while walking (WCostmin, J/m/kg). Gait features were extracted from the acceleration signal and included in a model to predict WCostmin.

RESULTS

On average WCostmin was 2.43±0.42 J/m/kg and correlated significantly with gait rate (r2 = 0.21, p<0.01) and regularity along the frontal (anteroposterior) and lateral (mediolateral) axes (r2 = 0.16, p<0.05 and r2 = 0.12, p<0.05 respectively). Together, the three variables explained 46% of the inter-subject variance (p<0.001) with a standard error of estimate of 0.30 J/m/kg. WCostmin and regularity along the frontal and lateral axes were related to age (WCostmin: r2 = 0.44, p<0.001; regularity: r2 = 0.16, p<0.05 and r2 = 0.12, p<0.05 respectively frontal and lateral).

CONCLUSIONS

The age associated decline in walking economy is induced by the adoption of an increased gait rate and by irregular body acceleration in the horizontal plane.

Self-adaptive fall-detection apparatus embedded in glasses

Fall injury is already a major problem in elderly health care. This work develops a self-adaptive fall-detection apparatus which is embedded in glasses for users easily to put on. The proposed system adopts a 9-axis sensing module of a triaxial magnetometer, accelerometer and gyroscope. First, the magnetometer is to filter out some normal events like head rotating, based on variations of rotation angles which are modeled by the Gaussian mixture model. Second, the sensed signals from a triaxial accelerometer are computed to obtain differential acceleration values at three directions, which are integrated and then compared with a threshold. Here, the threshold is determined by the Gaussian mixture model and optimized thresholding technique. Our system can update an adequate threshold on the fly. Third, when a fall occurs, its direction is identified using an accelerometer and a gyroscope. The experimental results reveal that the proposed system achieves accuracy rate of 92.1%, a specificity of 98.7%, and a sensitivity of 81.7%. As compared to the conventional fall-detection systems, the proposed system not only shows fairly good performance but also provides convenient, comfortable and non-intrusive wearing. Therefore, the system proposed herein can be widely spread in various head-mounted devices for health care applications.

Fall detection devices and their use with older adults: a systematic review

BACKGROUND

Falls represent a significant threat to the health and independence of adults aged 65 years and older. As a wide variety and large number of passive monitoring systems are currently and increasingly available to detect when individuals have fallen, there is a need to analyze and synthesize the evidence regarding their ability to accurately detect falls to determine which systems are most effective.

OBJECTIVES

The purpose of this literature review is to systematically assess the current state of design and implementation of fall-detection devices. This review also examines to what extent these devices have been tested in the real world as well as the acceptability of these devices to older adults.

DATA SOURCES

A systematic literature review was conducted in PubMed, CINAHL, EMBASE, and PsycINFO from their respective inception dates to June 25, 2013.

STUDY ELIGIBILITY CRITERIA AND INTERVENTIONS

Articles were included if they discussed a project or multiple projects involving a system with the purpose of detecting a fall in adults. It was not a requirement for inclusion in this review that the system targets persons older than 65 years. Articles were excluded if they were not written in English or if they looked at fall risk, fall detection in children, fall prevention, or a personal emergency response device.

STUDY APPRAISAL AND SYNTHESIS METHODS

Studies were initially divided into those using sensitivity, specificity, or accuracy in their evaluation methods and those using other methods to evaluate their devices. Studies were further classified into wearable devices and nonwearable devices. Studies were appraised for inclusion of older adults in sample and if evaluation included real-world settings.

RESULTS

This review identified 57 projects that used wearable systems and 35 projects using nonwearable systems, regardless of evaluation technique. Nonwearable systems included cameras, motion sensors, microphones, and floor sensors. Of the projects examining wearable systems, only 7.1% reported monitoring older adults in a real-world setting. There were no studies of nonwearable devices that used older adults as subjects in either a laboratory or a real-world setting. In general, older adults appear to be interested in using such devices although they express concerns over privacy and understanding exactly what the device is doing at specific times.

LIMITATIONS

This systematic review was limited to articles written in English and did not include gray literature. Manual paper screening and review processes may have been subject to interpretive bias.

CONCLUSIONS AND IMPLICATIONS OF KEY FINDINGS

There exists a large body of work describing various fall-detection devices. The challenge in this area is to create highly accurate unobtrusive devices. From this review it appears that the technology is becoming more able to accomplish such a task. There is a need now for more real-world tests as well as standardization of the evaluation of these devices.

Fall detection with body-worn sensors : a systematic review

BACKGROUND AND AIMS

Falls among older people remain a major public health challenge. Body-worn sensors are needed to improve the understanding of the underlying mechanisms and kinematics of falls. The aim of this systematic review is to assemble, extract and critically discuss the information available in published studies, as well as the characteristics of these investigations (fall documentation and technical characteristics).

METHODS

The searching of publically accessible electronic literature databases for articles on fall detection with body-worn sensors identified a collection of 96 records (33 journal articles, 60 conference proceedings and 3 project reports) published between 1998 and 2012. These publications were analysed by two independent expert reviewers. Information was extracted into a custom-built data form and processed using SPSS (SPSS Inc., Chicago, IL, USA).

RESULTS

The main findings were the lack of agreement between the methodology and documentation protocols (study, fall reporting and technical characteristics) used in the studies, as well as a substantial lack of real-world fall recordings. A methodological pitfall identified in most articles was the lack of an established fall definition. The types of sensors and their technical specifications varied considerably between studies.

CONCLUSION

Limited methodological agreement between sensor-based fall detection studies using body-worn sensors was identified. Published evidence-based support for commercially available fall detection devices is still lacking. A worldwide research group consensus is needed to address fundamental issues such as incident verification, the establishment of guidelines for fall reporting and the development of a common fall definition.