Fall-detection solution for mobile platforms using accelerometer and gyroscope data.

This paper defines two major data sets 1) from wearable inertial measurement sensors and 2) wearable ECG SHIMMER™ sensors. The first dataset is devised to benchmark techniques dealing with human behavior analysis based on multimodal inertial measurement wearable SHIMMER™ sensors unit during research studies “Fall Detection System for the Elderly Based on the Classification of Shimmer Sensor Prototype Data” [2] and “A novel fall detection algorithm for elderly using SHIMMER wearable sensors” [3]. The SHIMMER inertial sensor is a lightweight sensing device, incorporated with tri-axial accelerometer, a tri-axial gyroscope and tri-axial magnetometer, mounted on the waist of the subjects. The second dataset is developed to assess the feasibility of using SHIMMER™ wearable third generation ECG sensors for identification of basic heart anomalies by remote ECG analysis. The experimental protocol was carried out according to the Timed Up and Go (TUG) test [1], which is mainly used in fall detection and fall risk assessment systems specially designed for elderly. Three daily life activities such as standing still, walking and sitting on chair and getup were performed along with fall activity in controlled environment. This dataset is available on Data in Brief Dataverse [4] and a data repository [5].

As people get older, their bodies go through multiple changes that make them more fragile and susceptible to falls. The population of elderly people living alone is increasing worldwide, and this imposes a risk that a potential fall may happen without receiving prompt attention of a healthcare provider or caregiver. To solve this problem, various solutions for automatic fall detection have been proposed that recognise when a person falls and send alarms to someone that could provide quick help. One group of automatic fall detectors use wearable sensors attached to a person's body to measure body accelerations and then to distinguish falls from normal activities of daily living (ADLs) with some of the threshold or machine learning based algorithms. In threshold-based algorithms, features are calculated from measured accelerations and they are evaluated with a set of rules to check whether a fall has happened. The choice of fixed thresholds is thereby important for the overall efficiency of the algorithm. In our previous works in the field of fall detection, we have analysed methods for the determination of appropriate threshold levels for algorithms based on one acceleration-based feature. In this paper, we present a method for setting optimal thresholds in algorithms that use multiple acceleration-derived features. We demonstrate the efficiency of algorithms with thresholds set according to the newly presented method when tested on our dataset of accelerations measured during simulated falls and ADLs.

There has been a proliferation of smartphones, smart watches, and wearable sensors, making them ubiquitous in our daily lives. Mobile sensors have found widespread use due to their ever-decreasing cost, ease of deployment, and ability to provide continuous monitoring as opposed to sensors installed at fixed locations. Various techniques have been proposed for fall detection, gait analysis, activity monitoring, and heart rate and sleep sensing by wearable sensors and mobile phones. Compared to works that use inertial measurement unit (IMU) data or static cameras installed in the environment, there has been relatively less work using egocentric videos, meaning providing the first-person view from wearable cameras. Moreover, most of the existing studies on egocentric videos are based on only one sensor modality, namely the camera. There have been even fewer approaches that combine egocentric video data with IMU data. In this chapter, we will describe three different applications using wearable cameras together with IMU data. First, we will present an overview of a fall detection system using wearable devices, e.g., smartphones and tablets, equipped with cameras and accelerometers. Since the portable device is worn by the subject, monitoring is not limited to confined areas, and extends to wherever the subject may travel, as opposed to static sensors installed in certain rooms. Second, we will present an autonomous and robust method for counting footsteps, and tracking and calculating stride length by using both accelerometer and camera data from smartphones or Google™ glass. To provide higher precision, instead of using a preset stride length, the proposed method calculates the distance traveled with each step by using the camera data. This method is compared with the commercially available accelerometer-based step counter apps. The results show that the proposed method provides a significant increase in accuracy, and has the lowest average error rate both in number of steps taken and the distance traveled. Finally, we will provide an overview of a robust and autonomous method to detect activities with more details and context by using accelerometer and egocentric video data obtained from a smartphone.

As an essential subfield of context awareness, activity awareness, especially daily activity monitoring and fall detection, plays a significant role for elderly or frail people who need assistance in their daily activities. This study investigates the feature extraction and pattern recognition of surface electromyography (sEMG), with the purpose of determining the best features and classifiers of sEMG for daily living activities monitoring and fall detection. This is done by a serial of experiments. In the experiments, four channels of sEMG signal from wireless, wearable sensors located on lower limbs are recorded from three subjects while they perform seven activities of daily living (ADL). A simulated trip fall scenario is also considered with a custom-made device attached to the ankle. With this experimental setting, 15 feature extraction methods of sEMG, including time, frequency, time/frequency domain and entropy, are analyzed based on class separability and calculation complexity, and five classification methods, each with 15 features, are estimated with respect to the accuracy rate of recognition and calculation complexity for activity monitoring and fall detection. It is shown that a high accuracy rate of recognition and a minimal calculation time for daily activity monitoring and fall detection can be achieved in the current experimental setting. Specifically, the Wilson Amplitude (WAMP) feature performs the best, and the classifier Gaussian Kernel Support Vector Machine (GK-SVM) with Permutation Entropy (PE) or WAMP results in the highest accuracy for activity monitoring with recognition rates of 97.35% and 96.43%. For fall detection, the classifier Fuzzy Min-Max Neural Network (FMMNN) has the best sensitivity and specificity at the cost of the longest calculation time, while the classifier Gaussian Kernel Fisher Linear Discriminant Analysis (GK-FDA) with the feature WAMP guarantees a high sensitivity (98.70%) and specificity (98.59%) with a short calculation time (65.586 ms), making it a possible choice for pre-impact fall detection. The thorough quantitative comparison of the features and classifiers in this study supports the feasibility of a wireless, wearable sEMG sensor system for automatic activity monitoring and fall detection.

Background: Accidental falls are a significant health risk to older adults and patients seen in audiologyclinics. Personal emergency response systems are effective in preventing long lies (defined as remainingon the floor or ground for greater than one hour after a fall), but some individuals would prefer to wear less conspicuousdevices than a traditional neck-worn pendant. No previous investigation has compared theaccuracy of head-worn fall detection devices with those worn on other parts of the body. In this study, wecompared the accuracy of one commonly used fall detection pendant with two variants of a new hearinginstrument–based fall detection system.<br />Purpose: The purpose of the study was to evaluate the accuracy of detecting falls by using inertial sensorsembedded in hearing instruments.<br />Research Design: A study with repeated measures for one group.<br />Study Sample: Ten young adults served as participants. All participants had normal or corrected-to normalvision during testing. Individuals were excluded if they had self-reported cardiac disorders, recentconcussions, or musculoskeletal conditions.<br />Data Collection and Analysis: Data were collected from inertial measurement unit (IMU) sensors, embeddedinto a binaural set of hearing instruments, during laboratory-based simulations of falls, near-falls,and activities of daily living (ADLs). The detection state of a fall detection pendant was simultaneouslyrecorded during performance of each trial and compared with the outputs of offline hearing instrumentfirmware emulators.<br />Results: One hearing instrument–based fall detection system was more accurate than the fall detectionpendant. A second hearing instrument–based fall detection system performed similar to the fall detectionpendant. Each of the three fall detection systems investigated were robust against false-positive detectionsduring ADLs.<br />Conclusions: Preliminary findings demonstrate that hearing instruments embedded with a fall detectiondevice (FDD) may be a suitable alternative to more traditional forms of FDDs (e.g., pendant, wrist-worndevice, etc.) for detecting falls and potentially preventing long lies.<br />

The purpose of this study was to explore whether incorporating gyroscopic and accelerometer data will improve the prediction of energy expenditure (EE) of preschool children. Three model configurations were developed and compared using (1)accelerometer, (2)gyroscope, and (3)accelerometer + gyroscope data (dual sensor). Participants (n = 39; aged 3 to <6 years) were equipped with OPAL, GT9X, and GENEActiv devices, worn on the right hip, right wrist, and left wrist, while EE was simultaneously measured using a portable metabolic unit. The protocol consisted of semistructured activities spanning a range of intensities from low to high. A total of 54 machine learning models were developed to predict EE (2 EE measures [metabolic equivalents, kilojoules per minute] × 3 wear locations × 3 model types [random forest, linear regression, and fully connected neural network] × 3 sensor configurations). Model performance was evaluated using root mean squared error. Our findings reveal that, across the various configurations, the random forest model utilizing dual-sensor data achieved marginally lower mean root mean squared error in the majority of cases. Given the minimal improvements observed and the challenges associated with data acquisition, we recommend that researchers utilize accelerometer-based models moving forward.

As the population aging issue becomes more serious these days, fall detection of the elderly has been attracting a great deal of interests. A fall detection system includes data collection, data pre-processing, feature extraction, feature selection, and then activity classification. In this context the researchers have conducted studies on acceleration-based fall detection using external accelerometer or built-in sensors in smartphone. In this paper a novel approach for fall detection using machine learning technique is proposed, which employs a new pre-processing technique to remove noise from sensor data. It mainly consists of two processes: short-term smoothing to remove the short term vibration and long-term smoothing to smooth sensor readings captured in a longer time window. To detect falls, statistical models are proposed to extract the features. A public dataset, MobiFall, is used for performance evaluation, which contains the data of accelerometer and gyroscope with the orientation along each axis in the smartphone coordination system. With the selected features, the proposed scheme identifies falls from the activities of daily living with a high accuracy of up to 98.3%. Moreover, tFall dataset is also used to perform a cross verification of the proposed scheme.

Falls among the elderly are an important health issue which can lead to serious consequences. Unintentional falls in the elderly are the most common cause of nonfatal injuries. Different algorithms have been proposed for fall detection using vision based, ambient or inertial sensors. In this experiment, we used data collected from 16 subjects performing 15 tasks: 12 activities of daily living and 3 simulated falls. Subjects were equipped with a waist-mounted IMU to collect tri-axial accelerometer data during the activities. We derived features from accelerometer data on time-windowed identified activities and divided them into 2 different feature vectors (FV) that are used for fall detection algorithm. The selection was made based on the computational complexity of the features. We used different machine learning (ML) techniques to distinguish fall events from non-fall events (i.e. activities of daily living). In this paper, we compared and analyzed the accuracy of fall detection for different ML-based algorithms using the two accelerations-derived feature vectors for potential real-time implementation. The accuracy of the selected models ranged from 80.1 to 90.0% for FV1 and from 87.2 to 98.6% for FV2. The support vector machines algorithm achieved the best performance with accuracy 98.6%, sensitivity 98.7% and specificity 98.6% for FV2.

Falls are the main cause of injury in the elderly people. Therefore, fall detection system in the elderly is an essential technology for aging society. This project was designed to develop the fall detection system based on wearable sensors and Internet of Things for the real-time fall detection, data storage, calculation, and notification. The raw data, kinematics signal from the daily living of elder, including three-dimensional accelerometer and gyroscope are provided by GY-85, which have been sent bidots cloud via NodeMCU-ESP32. Further, the NodeMCU-ESP32 avail of the raw data to specify the fall detection algorithm. Moreover, the NodeMCU-ESP32 is designed to achieve the seamless communication between the fall detection device and system to acquire and process the data anywhere anytime as long as there is wifi connection. Consequently, the Ubidots cloud is designed to perform the calculation based on support vector machine (SVM) theory. It also sends the fall detection result via LINE notification (chat application) to caregivers. The tests of the proposed system are performed based on activities of daily living and the directions that fall can occur. The results of the testing show that the proposed system can provide high accuracy and reliability in fall detection.

A thorough analysis of sports is becoming increasingly important during the training process of badminton players at both the recreational and professional level. Nowadays, game situations are usually filmed and reviewed afterwards in order to analyze the game situation, but these video set-ups tend to be difficult to analyze, expensive, and intrusive to set up. In contrast, we classified badminton movements using off-the-shelf accelerometer and gyroscope data. To this end, we organized a data capturing campaign and designed a novel neural network using different frame sizes as input. This paper shows that with only accelerometer data, our novel convolutional neural network is able to distinguish nine activities with 86% precision when using a sampling frequency of 50 Hz. Adding the gyroscope data causes an increase of up to 99% precision, as compared to, respectively, 79% and 88% when using a traditional convolutional neural network. In addition, our paper analyses the impact of different sensor placement options and discusses the impact of different sampling frequenciess of the sensors. As such, our approach provides a low cost solution that is easy to use and can collect useful information for the analysis of a badminton game.

There are several large datasets available that are captured from motion tracking systems which could be useful to train wearable human activity recognition (HAR) systems, if only their spatial data could be mapped into the equivalent inertial measurement unit (IMU) data that would be sensed on the body. In this paper, we describe a mapping from 3D Vicon motion tracking data to data collected from a BlueSense on-body IMU. We characterise the error incurred in order to discern the extent to which it is possible to generate useful training data for a wearable activity recognition system from data collected with a motion capture system. We analyse this by mapping Vicon motion tracking data to rotational velocity and linear acceleration at the head, and compare this to actual gyroscope and accelerometer data collected by an IMU mounted on the head. In a 15 min dataset comprising three static activities—sitting, standing and lying down—we find that 95% of the reconstructed gyroscope data is within an error of [−7.25;+7.46] \(deg \cdot s^{-1}\), while 95% of the reconstructed accelerometer data was contained within [−96.1;+72.9] \(m \cdot G\). However, when we introduce more movement by including data collected while walking this increases to [−19.0;+18.2] \(deg \cdot s^{-1}\) for the gyroscope and [−208;+186] \(m \cdot G\) for the accelerometer. We conclude that generating accurate IMU data from motion capture datasets is possible and could be useful in providing larger volumes of data for activity recognition tasks and in helping enable advanced, data-hungry techniques such as deep learning to be employed on a larger scale within the domain of human activity recognition.

Smartphone Based Data Mining for Fall Detection: Analysis and Design

Fall risk assessment and fall detection are crucial for the prevention of adverse and long-term health outcomes. Wearable sensor systems have been used to assess fall risk and detect falls while providing additional meaningful information regarding gait characteristics. Commonly used wearable systems for this purpose are inertial measurement units (IMUs), which acquire data from accelerometers and gyroscopes. IMUs can be placed at various locations on the body to acquire motion data that can be further analyzed and interpreted. Insole-based devices are wearable systems that were also developed for fall risk assessment and fall detection. Insole-based systems are placed beneath the sole of the foot and typically obtain plantar pressure distribution data. Fall-related parameters have been investigated using inertial sensor-based and insole-based devices include, but are not limited to, center of pressure trajectory, postural stability, plantar pressure distribution and gait characteristics such as cadence, step length, single/double support ratio and stance/swing phase duration. The acquired data from inertial and insole-based systems can undergo various analysis techniques to provide meaningful information regarding an individual's fall risk or fall status. By assessing the merits and limitations of existing systems, future wearable sensors can be improved to allow for more accurate and convenient fall risk assessment. This article reviews inertial sensor-based and insole-based wearable devices that were developed for applications related to falls. This review identifies key points including spatiotemporal parameters, biomechanical gait parameters, physical activities and data analysis methods pertaining to recently developed systems, current challenges, and future perspectives.

This paper presents information on an autonomous fall detection and alerting system mainly consisting of a custom vest and a mobile smart phone. A 3-axial accelerator and gyroscope with a main board and Bluetooth are built into a custom vest worn by elderly. The vest can capture the data about the activities of daily living (ADLs) of elderly in real time. The data via Bluetooth is then sent to a mobile smart phone equipped with a threshold-based fall detection algorithm based on 3-axial accelerator and angular velocity. When a fall occurs the phone can alert a family member or health care center through a call or an emergent text message with a built in global positioning system (GPS) location, so as to provide a fast and accurate intervention. The experimental results show that the system discriminates falls from ADLs with a sensitivity of 95.74% and a specificity of 88.75%, for a total set of 175 movements.

Decreased physical activity in obese individuals is associated with a prevalence of cardiovascular and metabolic disorders. Physicians usually recommend that obese individuals change their lifestyle, specifically changes in diet, exercise, and other physical activities for obesity management. Therefore, understanding physical activity and sleep behavior is an essential aspect of obesity management. With innovations in mobile and electronic health care technologies, wearable inertial sensors have been used extensively over the past decade for monitoring human activities. Despite significant progress with the wearable inertial sensing technology, there is a knowledge gap among researchers regarding how to analyze longitudinal multi-day inertial sensor data to explore activities of daily living (ADL) and sleep behavior. The purpose of this study was to explore new clinically relevant metrics using movement amplitude and frequency from longitudinal wearable sensor data in obese and non-obese young adults. We utilized wavelet analysis to determine movement frequencies on longitudinal multi-day wearable sensor data. In this study, we recruited 10 obese and 10 non-obese young subjects. We found that obese participants performed more low-frequency (0.1 Hz) movements and fewer movements of high frequency (1.1–1.4 Hz) compared to non-obese counterparts. Both obese and non-obese subjects were active during the 00:00–06:00 time interval. In addition, obesity affected sleep with significantly fewer transitions, and obese individuals showed low values of root mean square transition accelerations throughout the night. This study is critical for obesity management to prevent unhealthy weight gain by the recommendations of physical activity based on our results. Longitudinal multi-day monitoring using wearable sensors has great potential to be integrated into routine health care checkups to prevent obesity and promote physical activities.