Fall detection--principles and methods

Radar-Based Fall Detection: A Survey

Fall detection, particularly critical for high-risk demographics like the elderly, is a key public health concern where timely detection can greatly minimize harm. With the advancements in radio frequency technology, radar has emerged as a powerful tool for human detection and tracking. Traditional machine learning algorithms, such as Support Vector Machines (SVM) and k-Nearest Neighbors (kNN), have shown promising outcomes. However, deep learning approaches, notably Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN), have outperformed in learning intricate features and managing large, unstructured datasets. This survey offers an in-depth analysis of radar-based fall detection, with emphasis on Micro-Doppler, Range-Doppler, and Range-Doppler-Angles techniques. We discuss the intricacies and challenges in fall detection and emphasize the necessity for a clear definition of falls and appropriate detection criteria, informed by diverse influencing factors. We present an overview of radar signal processing principles and the underlying technology of radar-based fall detection, providing an accessible insight into machine learning and deep learning algorithms. After examining 74 research articles on radar-based fall detection published since 2000, we aim to bridge current research gaps and underscore the potential future research strategies, emphasizing the real-world applications possibility and the unexplored potential of deep learning in improving radar-based fall detection.

A supervisory controller intended to arrest dynamic falls with a wearable cold-gas thruster

This article examines the feasibility of employing a cold-gas thruster (CGT), intended as a backpack-wearable device, for purposes of arresting backward falls, and in particular describes a supervisory controller that, for some motion described by an arbitrary combination of center-of-mass angle and angular velocity, both detects an impending fall and determines when to initiate thrust in the CGT in order to arrest the impending fall. The CGT prototype and the supervisory controller are described and experimentally assessed using a rocking block apparatus intended to approximate a backward-falling human. In these experiments, the CGT and supervisory controller restored upright stability to the rocking block in all experiment cases that would have otherwise resulted in a fall without the CGT assistance. Since the controller and experiments employ a reduced-order model of a falling human, the authors also conducted a series of simulations intended to examine the extent to which the controller might remain effective in the case of a multi-segment human. The results of these simulations suggest that the CGT controller would be nearly as effective on a multi-segment falling human as on the reduced-order model.

The Methods of Fall Detection: A Literature Review

Fall Detection Systems (FDS) are automated systems designed to detect falls experienced by older adults or individuals. Early or real-time detection of falls may reduce the risk of major problems. This literature review explores the current state of research on FDS and its applications. The review shows various types and strategies of fall detection methods. Each type of fall detection is discussed with its pros and cons. Datasets of fall detection systems are also discussed. Security and privacy issues related to fall detection systems are also considered in the discussion. The review also examines the challenges of fall detection methods. Sensors, algorithms, and validation methods related to fall detection are also talked over. This work found that fall detection research has gradually increased and become popular in the last four decades. The effectiveness and popularity of all strategies are also discussed. The literature review underscores the promising potential of FDS and highlights areas for further research and development.

Experimental Study: Deep Learning-Based Fall Monitoring among Older Adults with Skin-Wearable Electronics

Older adults are more vulnerable to falling due to normal changes due to aging, and their falls are a serious medical risk with high healthcare and societal costs. However, there is a lack of automatic fall detection systems for older adults. This paper reports (1) a wireless, flexible, skin-wearable electronic device for both accurate motion sensing and user comfort, and (2) a deep learning-based classification algorithm for reliable fall detection of older adults. The cost-effective skin-wearable motion monitoring device is designed and fabricated using thin copper films. It includes a six-axis motion sensor and is directly laminated on the skin without adhesives for the collection of accurate motion data. To study accurate fall detection using the proposed device, different deep learning models, body locations for the device placement, and input datasets are investigated using motion data based on various human activities. Our results indicate the optimal location to place the device is the chest, achieving accuracy of more than 98% for falls with motion data from older adults. Moreover, our results suggest a large motion dataset directly collected from older adults is essential to improve the accuracy of fall detection for the older adult population.

Feasibility Analysis of Using Channel State Information (CSI) Acquired from Wi-Fi Routers for Construction Worker Fall Detection

Accidental falls represent a major cause of fatal injuries for construction workers. Failure to seek medical attention after a fall can significantly increase the risk of death for construction workers. Wearable sensors, computer vision, and manual techniques are common modalities for detecting worker falls in the literature. However, they are severely constrained by issues such as cost, lighting, background, clutter, and privacy. To address the problems associated with the existing proposed methods, a new method has been conceived to identify construction worker falls by analyzing the CSI signals extracted from commercial Wi-Fi routers. In this research context, our study aimed to investigate the potential of using Channel State Information (CSI) to identify falls among construction workers. To achieve the aim of this study, CSI data corresponding to 360 sets of activities were collected from six construction workers on real construction sites. The results indicate that (1) the behavior of construction workers is highly correlated with the magnitude of CSI, even in real construction sites, and (2) the CSI-based method for identifying construction worker falls has an accuracy of 99% and can also accurately distinguish between falls and fall-like actions. The present study makes a significant contribution to the field by demonstrating the feasibility of utilizing low-cost Wi-Fi routers for the continuous monitoring of fall incidents among construction workers. To the best of our knowledge, this is the first investigation to address the issue of fall detection using commercial Wi-Fi devices in real-world construction environments. Considering the dynamic nature of construction sites, the new method developed in this study helps to detect falls at construction sites automatically and helps injured construction workers to seek medical attention on time.

Image and Speech Recognition Technology in the Development of an Elderly Care Robot: Practical Issues Review and Improvement Strategies

As the world's population is aging and there is a shortage of sufficient caring manpower, the development of intelligent care robots is a feasible solution. At present, plenty of care robots have been developed, but humanized care robots that can suitably respond to the individual behaviors of elderly people, such as pose, expression, gaze, and speech are generally lacking. To achieve the interaction, the main objectives of this study are: (1) conducting a literature review and analyzing the status quo on the following four core tasks of image and speech recognition technology: human pose recognition, human facial expression recognition, eye gazing recognition, and Chinese speech recognition; (2) proposing improvement strategies for these tasks based on the results of the literature review. The results of the study on these improvement strategies will provide the basis for using human facial expression robots in elderly care.

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.

Smart Wearables with Sensor Fusion for Fall Detection in Firefighting

During the past decade, falling has been one of the top three causes of death amongst firefighters in China. Even though there are many studies on fall-detection systems (FDSs), the majority use a single motion sensor. Furthermore, few existing studies have considered the impact sensor placement and positioning have on fall-detection performance; most are targeted toward fall detection of the elderly. Unfortunately, floor cracks and unstable building structures in the fireground increase the difficulty of detecting the fall of a firefighter. In particular, the movement activities of firefighters are more varied; hence, distinguishing fall-like activities from actual falls is a significant challenge. This study proposed a smart wearable FDS for firefighter fall detection by integrating motion sensors into the firefighter's personal protective clothing on the chest, elbows, wrists, thighs, and ankles. The firefighter's fall activities are detected by the proposed multisensory recurrent neural network, and the performances of different combinations of inertial measurement units (IMUs) on different body parts were also investigated. The results indicated that the sensor fusion of IMUs from all five proposed body parts achieved performances of 94.10%, 92.25%, and 94.59% in accuracy, sensitivity, and specificity, respectively.

The Stumblemeter: Design and Validation of a System That Detects and Classifies Stumbles during Gait

Stumbling during gait is commonly encountered in patients who suffer from mild to serious walking problems, e.g., after stroke, in osteoarthritis, or amputees using a lower leg prosthesis. Instead of self-reporting, an objective assessment of the number of stumbles in daily life would inform clinicians more accurately and enable the evaluation of treatments that aim to achieve a safer walking pattern. An easy-to-use wearable might fulfill this need. The goal of the present study was to investigate whether a single inertial measurement unit (IMU) placed at the shank and machine learning algorithms could be used to detect and classify stumbling events in a dataset comprising of a wide variety of daily movements. Ten healthy test subjects were deliberately tripped by an unexpected and unseen obstacle while walking on a treadmill. The subjects stumbled a total of 276 times, both using an elevating recovery strategy and a lowering recovery strategy. Subjects also performed multiple Activities of Daily Living. During data processing, an event-defined window segmentation technique was used to trace high peaks in acceleration that could potentially be stumbles. In the reduced dataset, time windows were labelled with the aid of video annotation. Subsequently, discriminative features were extracted and fed to train seven different types of machine learning algorithms. Trained machine learning algorithms were validated using leave-one-subject-out cross-validation. Support Vector Machine (SVM) algorithms were most successful, and could detect and classify stumbles with 100% sensitivity, 100% specificity, and 96.7% accuracy in the independent testing dataset. The SVM algorithms were implemented in a user-friendly, freely available, stumble detection app named Stumblemeter. This work shows that stumble detection and classification based on SVM is accurate and ready to apply in clinical practice.

REDECA: A Novel Framework to Review Artificial Intelligence and Its Applications in Occupational Safety and Health

Introduction: The field of artificial intelligence (AI) is rapidly expanding, with many applications seen routinely in health care, industry, and education, and increasingly in workplaces. Although there is growing evidence of applications of AI in workplaces across all industries to simplify and/or automate tasks there is a limited understanding of the role that AI contributes in addressing occupational safety and health (OSH) concerns. Methods: This paper introduces a new framework called Risk Evolution, Detection, Evaluation, and Control of Accidents (REDECA) that highlights the role that AI plays in the anticipation and control of exposure risks in a worker’s immediate environment. Two hundred and sixty AI papers across five sectors (oil and gas, mining, transportation, construction, and agriculture) were reviewed using the REDECA framework to highlight current applications and gaps in OSH and AI fields. Results: The REDECA framework highlighted the unique attributes and research focus of each of the five industrial sectors. The majority of evidence of AI in OSH research within the oil/gas and transportation sectors focused on the development of sensors to detect hazardous situations. In construction the focus was on the use of sensors to detect incidents. The research in the agriculture sector focused on sensors and actuators that removed workers from hazardous conditions. Application of the REDECA framework highlighted AI/OSH strengths and opportunities in various industries and potential areas for collaboration. Conclusions: As AI applications across industries continue to increase, further exploration of the benefits and challenges of AI applications in OSH is needed to optimally protect worker health, safety and well-being.

Man Down Situation Detection Using an in-Ear Inertial Platform

Man down situations (MDS) are a health or life threatening situations occurring largely in high-risk industrial workplaces. MDS automatic detection is crucial for workers safety especially in isolated working conditions where workers could be unable to call for help on their own, either due to loss of consciousness or an incapacitating injury. These solution must be reliable, robust, easy to use, but also have a low false-alarm rate, short response time and good ergonomics. This project aims to improve this technology by providing a global MDS definition according to a combination of three observable critical states based on characterization of body movement and orientation data from inertial measurements (accelerometer and gyroscope): the worker falls (F), worker immobility (I), the worker is down on the ground (D). The MDS detection strategy was established based on the detection of at least two distinct states, such as F-I, F-D or I-D, over a certain period of time. This strategy was tested using a large public database, revealing a significant reduction of the false alarms rate to 1.1%, reaching up to 99% accuracy. The proposed detection strategy was also incorporated into a digital earpiece, designed to address hearing protection issues, and validated according to an in vivo test procedure based on simulations of industrial workers normal activities and critical states.

A Machine Learning Multi-Class Approach for Fall Detection Systems Based on Wearable Sensors with a Study on Sampling Rates Selection

Falls are dangerous for the elderly, often causing serious injuries especially when the fallen person stays on the ground for a long time without assistance. This paper extends our previous work on the development of a Fall Detection System (FDS) using an inertial measurement unit worn at the waist. Data come from SisFall, a publicly available dataset containing records of Activities of Daily Living and falls. We first applied a preprocessing and a feature extraction stage before using five Machine Learning algorithms, allowing us to compare them. Ensemble learning algorithms such as Random Forest and Gradient Boosting have the best performance, with a Sensitivity and Specificity both close to 99%. Our contribution is: a multi-class classification approach for fall detection combined with a study of the effect of the sensors' sampling rate on the performance of the FDS. Our multi-class classification approach splits the fall into three phases: pre-fall, impact, post-fall. The extension to a multi-class problem is not trivial and we present a well-performing solution. We experimented sampling rates between 1 and 200 Hz. The results show that, while high sampling rates tend to improve performance, a sampling rate of 50 Hz is generally sufficient for an accurate detection.

A Simulator to Support Machine Learning-Based Wearable Fall Detection Systems

People’s life expectancy is increasing, resulting in a growing elderly population. That population is subject to dependency issues, falls being a problematic one due to the associated health complications. Some projects are trying to enhance the independence of elderly people by monitoring their status, typically by means of wearable devices. These devices often feature Machine Learning (ML) algorithms for fall detection using accelerometers. However, the software deployed often lacks reliable data for the models’ training. To overcome such an issue, we have developed a publicly available fall simulator capable of recreating accelerometer fall samples of two of the most common types of falls: syncope and forward. Those simulated samples are like real falls recorded using real accelerometers in order to use them later as input for ML applications. To validate our approach, we have used different classifiers over both simulated falls and data from two public datasets based on real data. Our tests show that the fall simulator achieves a high accuracy for generating accelerometer data from a fall, allowing to create larger datasets for training fall detection software in wearable devices.

Acceleration Magnitude at Impact Following Loss of Balance Can Be Estimated Using Deep Learning Model

Pre-impact fall detection can detect a fall before a body segment hits the ground. When it is integrated with a protective system, it can directly prevent an injury due to hitting the ground. An impact acceleration peak magnitude is one of key measurement factors that can affect the severity of an injury. It can be used as a design parameter for wearable protective devices to prevent injuries. In our study, a novel method is proposed to predict an impact acceleration magnitude after loss of balance using a single inertial measurement unit (IMU) sensor and a sequential-based deep learning model. Twenty-four healthy participants participated in this study for fall experiments. Each participant worn a single IMU sensor on the waist to collect tri-axial accelerometer and angular velocity data. A deep learning method, bi-directional long short-term memory (LSTM) regression, is applied to predict a fall's impact acceleration magnitude prior to fall impact (a fall in five directions). To improve prediction performance, a data augmentation technique with increment of dataset is applied. Our proposed model showed a mean absolute percentage error (MAPE) of 6.69 ± 0.33% with r value of 0.93 when all three different types of data augmentation techniques are applied. Additionally, there was a significant reduction of MAPE by 45.2% when the number of training datasets was increased by 4-fold. These results show that impact acceleration magnitude can be used as an activation parameter for fall prevention such as in a wearable airbag system by optimizing deployment process to minimize fall injury in real time.

IR-UWB Sensor Based Fall Detection Method Using CNN Algorithm

Falls are the leading cause of fatal injuries in the elderly such as fractures, and secondary damage from falls can lead to death. As such, fall detection is a crucial topic. However, due to the trade-off relationship between privacy preservation, user convenience, and fall detection performance, it is generally difficult to develop a fall detection system that simultaneously satisfies all conditions. The main goal of this study is to build a practical fall detection framework that can effectively classify the various behavior types into "Fall" and "Activities of daily living (ADL)" while securing privacy preservation and user convenience. For this purpose, signal data containing the motion information of objects was collected using a non-contact, unobtrusive, and non-restraint impulse-radio ultra wideband (IR-UWB) radar. These data were then applied to a convolutional neural network (CNN) algorithm to create an object behavior type classifier that can classify the behavior types of objects into "Fall" and "ADL." The data were collected by actually performing various activities of daily living, including falling. The performance of the classifier yielded satisfactory results. By combining an IR-UWB and CNN algorithm, this study demonstrates the feasibility of building a practical fall detection system that exceeds a certain level of detection accuracy while also ensuring privacy preservation and user convenience.

FPGA-based Edge Inferencing for Fall Detection

In the geriatric population, physical injuries sustained by an unintentional or an unpredictable fall on a hard surface is the leading cause of injury related morbidity and sometimes mortality. Each year, close to 30% of adults around the age group of 65 fall down at least once. In the year 2015, close to 2.9 million falls were reported, resulting in 33,000 deaths. As much as 61% of elderly nursing home residents fell at some point during their first year of residence.These falls may aggravate the situation leading to bone fracture, concussion, internal bleeding or traumatic brain injury when immediate medical attention is not offered to the person. Delay in course of the event may sometimes lead to death as well. Recently, many studies have come up with wearable devices. These devices that are now commercially available in the market are small, compact, wireless, battery operated and power efficient. This study discusses the findings that the optimal location for a Fall Detection Sensor on the human body is in front of the Shin bone. This is based on the 183 features collected from Inertial Measurement Unit (IMU) sensors placed on 16 human body locations and trained-tested using Convolutional Neural Networks (CNN) machine learning paradigm. The ultimate goal is to develop a mobile, wireless, wearable, low-power medical device that uses a small Lattice iCE40 Field Programmable Gate Array (FPGA) integrated with gyro and accelerometer sensors which detects whether the device wearer has fallen or not. This FPGA is capable of realizing the Neural Network model implemented in it. This Insitu or Edge inferencing wearable device is capable of providing real-time classifications without any Transmitting or Receiving capabilities over a wireless communication channel.

Cost of Cerebellar Ataxia in Hong Kong: A Retrospective Cost-of-Illness Analysis

Background: Cerebellar ataxia affects the coordination and balance of patients. The impact of this disease increases burden in patients, caregivers and society. Costs and the burden of this disease have not been investigated in Hong Kong.

Objectives: (1) To estimate the socioeconomic cost of cerebellar ataxia in Hong Kong for the base year 2019, (2) to assess the health-related quality of life (HRQoL) and severity of ataxia, and (3) to establish the correlation between the severity and cost of cerebellar ataxia and to examine the correlation between the severity of cerebellar ataxia and HRQoL.

Methods: A retrospective cross-sectional study was conducted amongst 31 patients with cerebellar ataxia. Cost-related data were obtained through self-reported questionnaires. The severity of ataxia was assessed using the Scale for Assessment and Rating of Ataxia, and HRQoL was assessed using the Short Form (36) Health Survey (SF-36). Pearson correlation was used for normally distributed data, whereas Spearman correlation was used otherwise.

Results: The mean severity of ataxia was 21 out of 40. The average direct and indirect costs of a patient with ataxia in 6 months were HKD 51,371 and HKD 93,855, respectively. The mean difference between the independent to minimally dependent in activities of daily living (ADL) group and the moderate to maximally dependent in ADL group for direct and indirect costs was HKD 33,829 and HKD 51,444, respectively. Significant expenditure was related to production lost (42%), caregiver salary (17%), and in-patient care (16%). The physical functioning (r = −0.58) and general health (r = −0.41) of SF-36 were negatively correlated with disease severity (p < 0.05). A significant, positive correlation was found between disease severity and direct cost (Spearman's rho = 0.39) and the cost of hiring a caregiver (Spearman's rho = 0.43).

Conclusion: The mean cost for 6 months for patients with cerebellar ataxia in Hong Kong is HKD 146,832. Additional support, including employment, access to specialist consultants, informal home care and community participation, are some areas that should be addressed. Future study on a larger population with a prospective design is necessary to confirm the aforementioned claims.

A Novel On-Wrist Fall Detection System Using Supervised Dictionary Learning Technique

Wrist-based fall detection system provides a very comfortable and multi-modal healthcare solution, especially for elderly risking falls. However, the wrist location presents a very challenging and unstable spot to distinguish falls among other daily activities. In this paper, we propose a Supervised Dictionary Learning approach for wrist-based fall detection. Three Dictionary learning algorithms for classification are invoked in this study, namely SRC, FDDL, and LRSDL. To extract the best descriptive representation of the signal data we followed different preprocessing scenarios based on accelerometer, gyroscope, and magnetometer. A considerable overall performance was obtained by the SRC algorithms reaching respectively 99.8%, 100%, and 96.6% of accuracy, sensitivity, and specificity using raw data provided by a triaxial accelerometer, accordingly overthrowing previously proposed methods for wrist placement.

A Pilot Study on Falling-Risk Detection Method Based on Postural Perturbation Evoked Potential Features

In the human-robot hybrid system, due to the error recognition of the pattern recognition system, the robot may perform erroneous motor execution, which may lead to falling-risk. While, the human can clearly detect the existence of errors, which is manifested in the central nervous activity characteristics. To date, the majority of studies on falling-risk detection have focused primarily on computer vision and physical signals. There are no reports of falling-risk detection methods based on neural activity. In this study, we propose a novel method to monitor multi erroneous motion events using electroencephalogram (EEG) features. There were 15 subjects who participated in this study, who kept standing with an upper limb supported posture and received an unpredictable postural perturbation. EEG signal analysis revealed a high negative peak with a maximum averaged amplitude of −14.75 ± 5.99 μV, occurring at 62 ms after postural perturbation. The xDAWN algorithm was used to reduce the high-dimension of EEG signal features. And, Bayesian linear discriminant analysis (BLDA) was used to train a classifier. The detection rate of the falling-risk onset is 98.67%. And the detection latency is 334ms, when we set detection rate beyond 90% as the standard of dangerous event onset. Further analysis showed that the falling-risk detection method based on postural perturbation evoked potential features has a good generalization ability. The model based on typical event data achieved 94.2% detection rate for unlearned atypical perturbation events. This study demonstrated the feasibility of using neural response to detect dangerous fall events.

Optimal Location for Fall Detection Edge Inferencing

A leading cause of physical injury sustained by elderly persons is the event of unintentionally falling onto a hard surface. Approximately 32-42% of those 70 and over fall at least once each year, and those who live in assisted living facilities fall with greater frequency per year than those who live in residential communities. Delay between the time of fall and the time of medical attention can exacerbate injury if the fall resulted in concussion, traumatic brain injury, or bone fracture. Several implementations of mobile, wireless, wearable, low-power fall detection sensors (FDS) have become commercially available. These devices are typically worn around the neck as a pendant, or on the wrist, as a watch is worn. Based on features collected from IMU sensors placed at sixteen body locations, and used to train four different machine learning models, our findings show the optimal placement for an FDS on the body is in front of the shinbone.

Robust Self-Adaptation Fall-Detection System Based on Camera Height

Vision-based fall-detection methods have been previously studied but many have limitations in terms of practicality. Due to differences in rooms, users do not set the camera or sensors at the same height. However, few studies have taken this into consideration. Moreover, some fall-detection methods are lacking in terms of practicality because only standing, sitting and falling are taken into account. Hence, this study constructs a data set consisting of various daily activities and fall events and studies the effect of camera/sensor height on fall-detection accuracy. Each activity in the data set is carried out by eight participants in eight directions and taken with the depth camera at five different heights. Many related studies heavily depended on human segmentation by using Kinect SDK but this is not reliable enough. To address this issue, this study proposes Enhanced Tracking and Denoising Alex-Net (ETDA-Net) to improve tracking and denoising performance and classify fall and non-fall events. Experimental results indicate that fall-detection accuracy is affected by camera height, against which ETDA-Net is robust, outperforming traditional deep learning based fall-detection methods.

A dataset for the development and optimization of fall detection algorithms based on wearable sensors

This paper describes a dataset acquired on 8 subjects while simulating 13 types of falls and 5 types of Activities of Daily Living (ADL), each repeated 3 times. In details, data includes 4 simulated falls forward (falling on knees ending up lying, ending in lateral position, ending up lying, ending up lying with recovery), 4 backward (falling sitting ending up lying, ending in lateral position, ending up lying, ending up lying with recovery), 2 lateral right (ending up lying, ending up lying with recovery), 2 lateral left (ending up lying, ending up lying with recovery), and 1 syncope. Simulated ADL are: lying on a bed then standing; walking a few meters; sitting on a chair then standing; go up or down three steps; and standing after picking something. Data were acquired using a MARG sensor, a wearable multisensory device tied to the subject's waist, that recorded time-variations of the subject's acceleration and orientation (expressed through the yaw, pitch and roll angles). These data can be useful in the development and test of algorithms to automatically identify and classify fall events. Fall detection systems are particularly useful when a subject is alone and not able to stand up after a fall, since an automatic alarm can be sent remotely to receive proper help.

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

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.

Machine learning-based pre-impact fall detection model to discriminate various types of fall

Preimpact fall detection can send alarm service faster to reduce long-lie conditions and decrease the risk of hospitalization. Detecting various types of fall to determine the impact site or direction prior to impact is important because it increases the chance of decreasing the incidence or severity of fall-related injuries. In this study, a robust preimpact fall detection model was developed to classify various activities and falls as multi-class and its performance was compared with the performance of previous developed models. Twelve healthy subjects participated in this study. All subjects were asked to place an inertial measuring unit module by fixing on a belt near the left iliac crest to collect accelerometer data for each activity. Our novel proposed model consists of feature calculation and infinite latent feature selection algorithm, auto labeling of activities, application of machine learning classifiers for discrete and continuous time series data. Nine machine-learning classifiers were applied to detect falls prior to impact and derive final detection results by sorting the classifier. Our model showed the highest classification accuracy. Results for the proposed model that could classify as multi-class showed significantly higher average classification accuracy of 99.57 ± 0.01% for discrete data-based classifiers and 99.84 ± 0.02% for continuous time series-based classifiers than previous models (p < 0.01). In the future, multi-class preimpact fall detection models can be applied to fall protector devices by detecting various activities for sending alerts or immediate feedback reactions to prevent falls.

Optimization and Technical Validation of the AIDE-MOI Fall Detection Algorithm in a Real-Life Setting with Older Adults

Falls are the primary contributors of accidents in elderly people. An important factor of fall severity is the amount of time that people lie on the ground. To minimize consequences through a short reaction time, the motion sensor "AIDE-MOI" was developed. "AIDE-MOI" senses acceleration data and analyzes if an event is a fall. The threshold-based fall detection algorithm was developed using motion data of young subjects collected in a lab setup. The aim of this study was to improve and validate the existing fall detection algorithm. In the two-phase study, twenty subjects (age 86.25 ± 6.66 years) with a high risk of fall (Morse > 65 points) were recruited to record motion data in real-time using the AIDE-MOI sensor. The data collected in the first phase (59 days) was used to optimize the existing algorithm. The optimized second-generation algorithm was evaluated in a second phase (66 days). The data collected in the two phases, which recorded 31 real falls, was split-up into one-minute chunks for labelling as "fall" or "non-fall". The sensitivity and specificity of the threshold-based algorithm improved significantly from 27.3% to 80.0% and 99.9957% (0.43) to 99.9978% (0.17 false alarms per week and subject), respectively.

Three-States-Transition Method for Fall Detection Algorithm Using Depth Image

Currently, the proportion of elderly persons is increasing all over the world, and accidents involving falls have become a serious problem especially for those who live alone. In this paper, an enhancement to our algorithm to detect such falls in an elderly person’s living room is proposed. Our previous algorithm obtains a binary image by using a depth camera and obtains an outline of the binary image by Canny edge detection. This algorithm then calculates the tangent vector angles of each outline pixels and divide them into 15° range groups. If most of the tangent angles are below 45°, a fall is detected. Traditional fall detection systems cannot detect falls towards the camera so at least two cameras are necessary in related works. To detect falls towards the camera, this study proposes the addition of a three-states-transition method to distinguish a fall state from a sitting-down one. The proposed algorithm computes the different position states and divides these states into three groups to detect the person’s current state. Futhermore, transition speed is calculated in order to differentiate sit states from fall states. This study constructes a data set that includes over 1500 images, and the experimental evaluation of the images demonstrates that our enhanced algorithm is effective for detecting the falls with only a single camera.

Transfer learning approach for fall detection with the FARSEEING real-world dataset and simulated falls

Falls are very rare and extremely difficult to acquire in free living conditions. Due to this, most of prior work on fall detection has focused on simulated datasets acquired in scenarios that mimic the real-world context, however, the validation of systems trained with simulated falls remains unclear. This work presents a transfer learning approach for combining a dataset of simulated falls and non-falls, obtained from young volunteers, with the real-world FARSEEING dataset, in order to train a set of supervised classifiers for discriminating between falls and non-falls events. The objective is to analyze if a combination of simulated and real falls could enrich the model. In the real-world, falls are a sporadic event, which results in imbalanced datasets. In this work, several methods for imbalance learning were employed: SMOTE, Balance Cascade and Ranking models. The Balance Cascade obtained less misclassifications in the validation set.There was an improvement when mixing the real falls and simulated non-falls compared to the case when only simulated falls were used for training. When testing with a mixed set with real falls and simulated non-falls, it is even more important to train with a mixed set. Moreover, it was possible to onclude that a model trained with simulated falls generalize better when tested with real falls, than the opposite. The overall accuracy obtained for the combination of different datasets were above 95 %.

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

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.

An Event-Triggered Machine Learning Approach for Accelerometer-Based Fall Detection

The fixed-size non-overlapping sliding window (FNSW) and fixed-size overlapping sliding window (FOSW) approaches are the most commonly used data-segmentation techniques in machine learning-based fall detection using accelerometer sensors. However, these techniques do not segment by fall stages (pre-impact, impact, and post-impact) and thus useful information is lost, which may reduce the detection rate of the classifier. Aligning the segment with the fall stage is difficult, as the segment size varies. We propose an event-triggered machine learning (EvenT-ML) approach that aligns each fall stage so that the characteristic features of the fall stages are more easily recognized. To evaluate our approach, two publicly accessible datasets were used. Classification and regression tree (CART), k-nearest neighbor (k-NN), logistic regression (LR), and the support vector machine (SVM) were used to train the classifiers. EvenT-ML gives classifier F-scores of 98% for a chest-worn sensor and 92% for a waist-worn sensor, and significantly reduces the computational cost compared with the FNSW- and FOSW-based approaches, with reductions of up to 8-fold and 78-fold, respectively. EvenT-ML achieves a significantly better F-score than existing fall detection approaches. These results indicate that aligning feature segments with fall stages significantly increases the detection rate and reduces the computational cost.

A Radar-Based Smart Sensor for Unobtrusive Elderly Monitoring in Ambient Assisted Living Applications

Continuous in-home monitoring of older adults living alone aims to improve their quality of life and independence, by detecting early signs of illness and functional decline or emergency conditions. To meet requirements for technology acceptance by seniors (unobtrusiveness, non-intrusiveness, and privacy-preservation), this study presents and discusses a new smart sensor system for the detection of abnormalities during daily activities, based on ultra-wideband radar providing rich, not privacy-sensitive, information useful for sensing both cardiorespiratory and body movements, regardless of ambient lighting conditions and physical obstructions (through-wall sensing). The radar sensing is a very promising technology, enabling the measurement of vital signs and body movements at a distance, and thus meeting both requirements of unobtrusiveness and accuracy. In particular, impulse-radio ultra-wideband radar has attracted considerable attention in recent years thanks to many properties that make it useful for assisted living purposes. The proposed sensing system, evaluated in meaningful assisted living scenarios by involving 30 participants, exhibited the ability to detect vital signs, to discriminate among dangerous situations and activities of daily living, and to accommodate individual physical characteristics and habits. The reported results show that vital signs can be detected also while carrying out daily activities or after a fall event (post-fall phase), with accuracy varying according to the level of movements, reaching up to 95% and 91% in detecting respiration and heart rates, respectively. Similarly, good results were achieved in fall detection by using the micro-motion signature and unsupervised learning, with sensitivity and specificity greater than 97% and 90%, respectively.

HDOMO: Smart Sensor Integration for an Active and Independent Longevity of the Elderly

The aim of this paper is to present the main results of HDOMO, an Ambient Assisted Living (AAL) project that involved 16 Small and Medium Enterprises (SMEs) and 2 research institutes. The objective of the project was to create an autonomous and automated domestic environment, primarily for elderly people and people with physical and motor disabilities. A known and familiar environment should help users in their daily activities and it should act as a virtual caregiver by calling, if necessary, relief efforts. Substantially, the aim of the project is to simplify the life of people in need of support, while keeping them autonomous in their private environment. From a technical point of view, the project provides the use of different Smart Objects (SOs), able to communicate among each other, in a cloud base infrastructure, and with the assisted users and their caregivers, in a perspective of interoperability and standardization of devices, usability and effectiveness of alarm systems. In the state of the art there are projects that achieve only a few of the elements listed. The HDOMO project aims to achieve all of them in one single project effectively. The experimental trials performed in a real scenario demonstrated the accuracy and efficiency of the system in extracting and processing data in real time to promptly acting, and in providing timely response to the needs of the user by integrating and confirming main alarms with different interoperable smart sensors. The article proposes a new technique to improve the accuracy of the system in detecting alarms using a multi-SO approach with information fusion between different devices, proving that this architecture can provide robust and reliable results on real environments.

An Energy-Efficient Multi-Tier Architecture for Fall Detection Using Smartphones

Automatic detection of fall events is vital to providing fast medical assistance to the causality, particularly when the injury causes loss of consciousness. Optimization of the energy consumption of mobile applications, especially those which run 24/7 in the background, is essential for longer use of smartphones. In order to improve energy-efficiency without compromising on the fall detection performance, we propose a novel 3-tier architecture that combines simple thresholding methods with machine learning algorithms. The proposed method is implemented on a mobile application, called uSurvive, for Android smartphones. It runs as a background service and monitors the activities of a person in daily life and automatically sends a notification to the appropriate authorities and/or user defined contacts when it detects a fall. The performance of the proposed method was evaluated in terms of fall detection performance and energy consumption. Real life performance tests conducted on two different models of smartphone demonstrate that our 3-tier architecture with feature reduction could save up to 62% of energy compared to machine learning only solutions. In addition to this energy saving, the hybrid method has a 93% of accuracy, which is superior to thresholding methods and better than machine learning only solutions.

A Combined One-Class SVM and Template-Matching Approach for User-Aided Human Fall Detection by Means of Floor Acoustic Features

The primary cause of injury-related death for the elders is represented by falls. The scientific community devoted them particular attention, since injuries can be limited by an early detection of the event. The solution proposed in this paper is based on a combined One-Class SVM (OCSVM) and template-matching classifier that discriminate human falls from nonfalls in a semisupervised framework. Acoustic signals are captured by means of a Floor Acoustic Sensor; then Mel-Frequency Cepstral Coefficients and Gaussian Mean Supervectors (GMSs) are extracted for the fall/nonfall discrimination. Here we propose a single-sensor two-stage user-aided approach: in the first stage, the OCSVM detects abnormal acoustic events. In the second, the template-matching classifier produces the final decision exploiting a set of template GMSs related to the events marked as false positives by the user. The performance of the algorithm has been evaluated on a corpus containing human falls and nonfall sounds. Compared to the OCSVM only approach, the proposed algorithm improves the performance by 10.14% in clean conditions and 4.84% in noisy conditions. Compared to Popescu and Mahnot (2009) the performance improvement is 19.96% in clean conditions and 8.08% in noisy conditions.

Real-time measures of context to improve fall-detection models

Real-time fall detection has been a challenging area of research and even more challenging as a viable commercial service, given the need for near perfect classification algorithms. True fall events are rare is monitored data sets, whereas confounding events for automated algorithms are quite frequent. In this paper we describe a decision theoretic approach to classification and alerting that incorporates context, such as location and activities, to improve probability and utility estimates for new classes, including near falls and known confounding events. We describe how to use monitored context to provide real-time assessment of true patient state to improve training data sets, as well as the use of context in improving classification, detection and alerting.

Fall risk probability estimation based on supervised feature learning using public fall datasets

Risk of falling is considered among major threats for elderly population and therefore started to play an important role in modern healthcare. With recent development of sensor technology, the number of studies dedicated to reliable fall detection system has increased drastically. However, there is still a lack of universal approach regarding the evaluation of developed algorithms. In the following study we make an attempt to find publicly available fall datasets and analyze similarities among them using supervised learning. After preforming similarity assessment based on multidimensional scaling we indicate the most representative feature vector corresponding to each specific dataset. This vector obtained from a real-life data is subsequently deployed to estimate fall risk probabilities for a statistical fall detection model. Finally, we conclude with some observations regarding the similarity assessment results and provide suggestions towards an efficient approach for evaluation of fall detection studies.

Novel Hierarchical Fall Detection Algorithm Using a Multiphase Fall Model

Falls are the primary cause of accidents for the elderly in the living environment. Reducing hazards in the living environment and performing exercises for training balance and muscles are the common strategies for fall prevention. However, falls cannot be avoided completely; fall detection provides an alarm that can decrease injuries or death caused by the lack of rescue. The automatic fall detection system has opportunities to provide real-time emergency alarms for improving the safety and quality of home healthcare services. Two common technical challenges are also tackled in order to provide a reliable fall detection algorithm, including variability and ambiguity. We propose a novel hierarchical fall detection algorithm involving threshold-based and knowledge-based approaches to detect a fall event. The threshold-based approach efficiently supports the detection and identification of fall events from continuous sensor data. A multiphase fall model is utilized, including free fall, impact, and rest phases for the knowledge-based approach, which identifies fall events and has the potential to deal with the aforementioned technical challenges of a fall detection system. Seven kinds of falls and seven types of daily activities arranged in an experiment are used to explore the performance of the proposed fall detection algorithm. The overall performances of the sensitivity, specificity, precision, and accuracy using a knowledge-based algorithm are 99.79%, 98.74%, 99.05% and 99.33%, respectively. The results show that the proposed novel hierarchical fall detection algorithm can cope with the variability and ambiguity of the technical challenges and fulfill the reliability, adaptability, and flexibility requirements of an automatic fall detection system with respect to the individual differences.

SisFall: A Fall and Movement Dataset

Research on fall and movement detection with wearable devices has witnessed promising growth. However, there are few publicly available datasets, all recorded with smartphones, which are insufficient for testing new proposals due to their absence of objective population, lack of performed activities, and limited information. Here, we present a dataset of falls and activities of daily living (ADLs) acquired with a self-developed device composed of two types of accelerometer and one gyroscope. It consists of 19 ADLs and 15 fall types performed by 23 young adults, 15 ADL types performed by 14 healthy and independent participants over 62 years old, and data from one participant of 60 years old that performed all ADLs and falls. These activities were selected based on a survey and a literature analysis. We test the dataset with widely used feature extraction and a simple to implement threshold based classification, achieving up to 96% of accuracy in fall detection. An individual activity analysis demonstrates that most errors coincide in a few number of activities where new approaches could be focused. Finally, validation tests with elderly people significantly reduced the fall detection performance of the tested features. This validates findings of other authors and encourages developing new strategies with this new dataset as the benchmark.

Two-threshold energy based fall detection using a triaxial accelerometer

Elderly fall detection based on accelerometers is an active research area. Nowadays authors are addressing specific problems such as failure rates and energy consumption, but in most cases their strategies do not conciliate these objectives. In this paper we propose a double threshold based methodology with two novel detection features, a product between the sum vector magnitude and the signal magnitude area, and a normalization of the signal magnitude area over five 1 s windows. The methodology was validated using the public Mobifall dataset, and one developed for this work. It achieved 99 % of accuracy with Mobifall, and 97 % with the self-developed dataset. This methodology is based on an activity by activity analysis performed for determining which activities are prone to fail, as an alternative way of reducing detection failures.

An Analysis on Sensor Locations of the Human Body for Wearable Fall Detection Devices: Principles and Practice

Wearable devices for fall detection have received attention in academia and industry, because falls are very dangerous, especially for elderly people, and if immediate aid is not provided, it may result in death. However, some predictive devices are not easily worn by elderly people. In this work, a huge dataset, including 2520 tests, is employed to determine the best sensor placement location on the body and to reduce the number of sensor nodes for device ergonomics. During the tests, the volunteer’s movements are recorded with six groups of sensors each with a triaxial (accelerometer, gyroscope and magnetometer) sensor, which is placed tightly on different parts of the body with special straps: head, chest, waist, right-wrist, right-thigh and right-ankle. The accuracy of individual sensor groups with their location is investigated with six machine learning techniques, namely the k-nearest neighbor (k-NN) classifier, Bayesian decision making (BDM), support vector machines (SVM), least squares method (LSM), dynamic time warping (DTW) and artificial neural networks (ANNs). Each technique is applied to single, double, triple, quadruple, quintuple and sextuple sensor configurations. These configurations create 63 different combinations, and for six machine learning techniques, a total of 63 × 6 = 378 combinations is investigated. As a result, the waist region is found to be the most suitable location for sensor placement on the body with 99.96% fall detection sensitivity by using the k-NN classifier, whereas the best sensitivity achieved by the wrist sensor is 97.37%, despite this location being highly preferred for today’s wearable applications.

Older Adults' Experience with a Novel Fall Detection Device

BACKGROUND

Falls are a significant concern for the older adult (OA) population, many of whom are unable to get up following a fall.

INTRODUCTION

While many devices exist designed to detect a fall, little work has been conducted to evaluate the usability of such devices. We present a longitudinal usability study of a fall detection (FD) device tested with OAs in real-world settings.

MATERIALS AND METHODS

OAs were recruited and asked to use a wearable FD device for up to 4 months. Participants were interviewed at baseline and 2 and 4 months and encouraged to provide direct feedback on their experience.

RESULTS

In total, 18 OAs participated in the study. Eight completed the 4-month trial. We conducted a total of 38 interviews (16 baseline, 7 midpoint, and 15 final) and logged a total of 78 comments. While participants enjoyed the GPS and automatic detection features of the device, they were unhappy with the volume of false alarms and obtrusiveness of the device. Many also did not see a great need for having the device or were embarrassed by the device.

DISCUSSION

Engineers must work to better develop this technology so that it is accessible to people with hearing loss, limited dexterity, and low vision. Utilizing age-appropriate design techniques will help make such informatics tools more user friendly.

CONCLUSION

We explored the usability of a particular FD device with OAs and provide design recommendations to help future device manufacturers create more age-appropriate devices.

Smart walkers: an application-oriented review

In this paper, a development method for smart walker prototypes is proposed. Development of such prototypes is based on technological choices and device evaluations. The method is aimed at guiding technological choices in a modular fashion. First, the method for choosing modules to be integrated in a smart walker is presented. Application-specific modules are then studied. Finally, the issues of evaluation are investigated. In order to work out this method, more than 50 smart walkers and their pros and cons with respect to the different studied applications are reviewed.