A wearable system for pre-impact fall detection

Enhancing Patient Safety Through an Integrated Internet of Things Patient Care System: Large Quasi-Experimental Study on Fall Prevention

BACKGROUND

The challenge of preventing in-patient falls remains one of the most critical concerns in health care.

OBJECTIVE

This study aims to investigate the effect of an integrated Internet of Things (IoT) smart patient care system on fall prevention.

METHODS

A quasi-experimental study design is used. The smart patient care system is an integrated IoT system combining a motion-sensing mattress for bed-exit detection, specifying different types of patient calls, integrating a health care staff scheduling system, and allowing health care staff to receive and respond to alarms via mobile devices. Unadjusted and adjusted logistic regression models were used to investigate the relationship between the use of the IoT system and bedside falls compared with a traditional patient care system.

RESULTS

In total, 1300 patients were recruited from a medical center in Taiwan. The IoT patient care system detected an average of 13.5 potential falls per day without any false alarms, whereas the traditional system issued about 11 bed-exit alarms daily, with approximately 4 being false, effectively identifying 7 potential falls. The bedside fall incidence during hospitalization was 1.2% (n=8) in the traditional patient care system ward and 0.1% (n=1) in the smart ward. We found that the likelihood of bedside falls in wards with the IoT system was reduced by 88% (odds ratio 0.12, 95% CI 0.01-0.97; P=.047).

CONCLUSIONS

The integrated IoT smart patient care system might prevent falls by assisting health care staff with efficient and resilient responses to bed-exit detection. Future product development and research are recommended to introduce IoT into patient care systems combining bed-exit alerts to prevent inpatient falls and address challenges in patient safety.

Acquisition of Data on Kinematic Responses to Unpredictable Gait Perturbations: Collection and Quality Assurance of Data for Use in Machine Learning Algorithms for (Near-)Fall Detection

Slip, trip, and fall (STF) accidents cause high rates of absence from work in many companies. During the 2022 reporting period, the German Social Accident Insurance recorded 165,420 STF accidents, of which 12 were fatal and 2485 led to disability pensions. Particularly in the traffic, transport and logistics sector, STF accidents are the most frequently reported occupational accidents. Therefore, an accurate detection of near-falls is critical to improve worker safety. Efficient detection algorithms are essential for this, but their performance heavily depends on large, well-curated datasets. However, there are drawbacks to current datasets, including small sample sizes, an emphasis on older demographics, and a reliance on simulated rather than real data. In this paper we report the collection of a standardised kinematic STF dataset from real-world STF events affecting parcel delivery workers and steelworkers. We further discuss the use of the data to evaluate dynamic stability control during locomotion for machine learning and build a standardised database. We present the data collection, discuss the classification of the data, present the totality of the data statistically, and compare it with existing databases. A significant research gap is the limited number of participants and focus on older populations in previous studies, as well as the reliance on simulated rather than real-world data. Our study addresses these gaps by providing a larger dataset of real-world STF events from a working population with physically demanding jobs. The population studied included 110 participants, consisting of 55 parcel delivery drivers and 55 steelworkers, both male and female, aged between 19 and 63 years. This diverse participant base allows for a more comprehensive understanding of STF incidents in different working environments.

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.

Fall Direction Detection in Motion State Based on the FMCW Radar

Accurately detecting falls and providing clear directions for the fall can greatly assist medical staff in promptly developing rescue plans and reducing secondary injuries during transportation to the hospital. In order to facilitate portability and protect people's privacy, this paper presents a novel method for detecting fall direction during motion using the FMCW radar. We analyze the fall direction in motion based on the correlation between different motion states. The range-time (RT) features and Doppler-time (DT) features of the person from the motion state to the fallen state were obtained by using the FMCW radar. We analyzed the different features of the two states and used a two-branch convolutional neural network (CNN) to detect the falling direction of the person. In order to improve the reliability of the model, this paper presents a pattern feature extraction (PFE) algorithm that effectively eliminates noise and outliers in RT maps and DT maps. The experimental results show that the method proposed in this paper has an identification accuracy of 96.27% for different falling directions, which can accurately identify the falling direction and improve the efficiency of rescue.

Improving Fall Detection Devices for Older Adults Using Quality Function Deployment (QFD) Approach

Engineering invention must be in tandem with public demands. Often it is difficult to identify the priorities of consumers where technological advancement is needed. In line with the global challenge of increasing fall prevalence among older adults, providing prevention solutions is the key. This study aims at developing an improved fall detection device using an approach called Quality Function Deployment (QFD). The goal is to investigate features to incorporate in existing device from consumer's perspectives. A three-phases design process is constructed; (1) Questionnaire, (2) Ishikawa Method, and (3) QFD. The proposed method begins with identifying customer needs as the requirement analysis, followed by a method to convert them to design specifications to be added in a fall detection device using QFD tool. As the top feature is monitoring balance, the new improved fall detection devices incorporating balance features will help older adults to monitor their level of risk of falling.

Feature-Based Information Retrieval of Multimodal Biosignals with a Self-Similarity Matrix: Focus on Automatic Segmentation

Biosignal-based technology has been increasingly available in our daily life, being a critical information source. Wearable biosensors have been widely applied in, among others, biometrics, sports, health care, rehabilitation assistance, and edutainment. Continuous data collection from biodevices provides a valuable volume of information, which needs to be curated and prepared before serving machine learning applications. One of the universal preparation steps is data segmentation and labelling/annotation. This work proposes a practical and manageable way to automatically segment and label single-channel or multimodal biosignal data using a self-similarity matrix (SSM) computed with signals' feature-based representation. Applied to public biosignal datasets and a benchmark for change point detection, the proposed approach delivered lucid visual support in interpreting the biosignals with the SSM while performing accurate automatic segmentation of biosignals with the help of the novelty function and associating the segments grounded on their similarity measures with the similarity profiles. The proposed method performed superior to other algorithms in most cases of a series of automatic biosignal segmentation tasks; of equal appeal is that it provides an intuitive visualization for information retrieval of multimodal biosignals.

Deep Learning-Based Near-Fall Detection Algorithm for Fall Risk Monitoring System Using a Single Inertial Measurement Unit

Proactively detecting falls and preventing injuries are among the primary keys to a healthy life for the elderly. Near-fall remote monitoring in daily life could provide key information to prevent future falls and obtain quantitative rehabilitation status for patients with weak balance ability. In this study, we developed a deep learning-based novel classification algorithm to precisely categorize three classes (falls, near-falls, and activities of daily living (ADLs)) using a single inertial measurement unit (IMU) device attached to the waist. A total of 34 young participants were included in this study. An IMU containing accelerometer and gyroscope sensors was fabricated to acquire acceleration and angular velocity signals. A comprehensive experiment including thirty-six types of activities (10 types of falls, 10 types of near-falls, and 16 types of ADLs) was designed based on previous studies. A modified directed acyclic graph-convolution neural network (DAG-CNN) architecture with hyperparameter optimization was proposed to predict fall, near-fall, and ADLs. Prediction results of the modified DAG-CNN structure were found to be approximately 7% more accurate than the traditional CNN structure. For the case of near-falls, the modified DAG-CNN demonstrated excellent prediction performance with accuracy of over 98% by combining gyroscope and accelerometer features. Additionally, by combining acceleration and angular velocity the trained model showed better performance than each model of acceleration and angular velocity. It is believed that information to preemptively handle the risk of falls and quantitatively evaluate the rehabilitation status of the elderly with weak balance will be provided by monitoring near-falls.

Human Activity Recognition Based on Non-Contact Radar Data and Improved PCA Method

Human activity recognition (HAR) can effectively improve the safety of the elderly at home. However, non-contact millimeter-wave radar data on the activities of the elderly is often challenging to collect, making it difficult to effectively improve the accuracy of neural networks for HAR. We addressed this problem by proposing a method that combines the improved principal component analysis (PCA) and the improved VGG16 model (a pre-trained 16-layer neural network model) to enhance the accuracy of HAR under small-scale datasets. This method used the improved PCA to enhance features of the extracted components and reduce the dimensionality of the data. The VGG16 model was improved by deleting the complex Fully-Connected layers and adding a Dropout layer between them to prevent the loss of useful information. The experimental results show that the accuracy of our proposed method on HAR is 96.34%, which is 4.27% higher after improvement, and the training time of each round is 10.88 s, which is 12.8% shorter than before.

Health technologies for the prevention and detection of falls in adult hospital inpatients: a scoping review

OBJECTIVE

The objective of this scoping review was to examine and map the evidence relating to the reporting and evaluation of technologies for the prevention and detection of falls in adult hospital inpatients.

INTRODUCTION

Falls are a common cause of accidental injury, leading to significant safety issues in hospitals globally, and resulting in substantial human and economic costs. Previous research has focused on community settings with less emphasis on hospital settings.

INCLUSION CRITERIA

Participants included adult inpatients, aged 18 years and over; the concept included the use of fall-prevention or fall-detection technologies; the context included any hospital ward setting.

METHODS

This scoping review was conducted according to JBI methodology for scoping reviews, guided by an a priori protocol. A wide selection of databases including MEDLINE, CINAHL, AMED, Embase, PEDro, Epistimonikos, and Science Direct were searched for records from inception to October 2019. Other sources included gray literature, trial registers, government health department websites, and websites of professional bodies. Only studies in the English language were included. A three-step search strategy was employed, with all records exported for subsequent title and abstract screening prior to full-text screening. Screening was performed by two independent reviewers and data extraction by one reviewer following agreement checks. Data are presented in narrative and tabular form.

RESULTS

Over 13,000 records were identified with 404 included in the scoping review: 336 reported on fall-prevention technologies, 51 targeted detection, and 17 concerned both. The largest contributions of studies came from the USA (n=185), Australia (n=65), the UK (n=36), and Canada (n=18). There was a variety of study designs including 77 prospective cohort studies, 33 before-after studies, and 35 systematic reviews; however, relatively few randomized controlled trials were conducted (n = 25). The majority of records reported on multifactorial and multicomponent technologies (n = 178), followed by fall detection devices (n = 86). Few studies reported on the following interventions in isolation: fall risk assessment (n = 6), environment design (n = 8), sitters (n = 5), rounding (n = 3), exercise (n = 3), medical/pharmaceutical (n = 2), physiotherapy (n = 1), and nutritional (n = 1). The majority (57%) of studies reported clinical effectiveness outcomes, with smaller numbers (14%) reporting feasibility and/or acceptability outcomes, or cost-effectiveness outcomes (5%).

CONCLUSIONS

This review has mapped the literature on fall-prevention and fall-detection technology and outcomes for adults in the hospital setting. Despite the volume of available literature, there remains a need for further high-quality research on fall-prevention and fall-detection technologies.

Damaged/missing proximity sensor induces screen mistouch when answering calls: Prediction of smartphone answering status by posture data

When the proximity sensor of a smartphone is impaired, it would easily lead to screen mistouch during conversation, which will significantly affect the user experience. However, there are relatively few studies that have been focused on the quality of user experience following sensor impairment. The purpose of this study was to compare and evaluate different machine learning models in forecasting the user’s posture during a phone call, thereby providing a compensation approach for detecting proximity to the human ear during a phone call following sensor damage. The built-in accelerometer sensors of smartphones were employed to collect posture data while users were employing their smartphones. Three main postures (holding, moving and answering) were identified; the posture data were obtained through training and prediction using five machine learning models. The results showed that the model that utilized triaxial data had better prediction accuracy than the model that used single-axis data. Furthermore, models with time-domain features had a higher accuracy rate. Among the five models, neural networks had the best prediction accuracy (0.982). The proposed approach could be of immense benefit to the users following proximity sensor damage, and would be advantageous in the design of the smartphone, particularly in the early stages of the design process.

A smartphone-based online system for fall detection with alert notifications and contextual information of real-life falls

BACKGROUND

Falls are a leading cause of accidental deaths and injuries worldwide. The risk of falling is especially high for individuals suffering from balance impairments. Retrospective surveys and studies of simulated falling in lab conditions are frequently used and are informative, but prospective information about real-life falls remains sparse. Such data are essential to address fall risks and develop fall detection and alert systems. Here we present the results of a prospective study investigating a proof-of-concept, smartphone-based, online system for fall detection and notification.

METHODS

The system uses the smartphone's accelerometer and gyroscope to monitor the participants' motion, and falls are detected using a regularized logistic regression. Data on falls and near-fall events (i.e., stumbles) is stored in a cloud server and fall-related variables are logged onto a web portal developed for data exploration, including the event time and weather, fall probability, and the faller's location and activity before the fall.

RESULTS

In total, 23 individuals with an elevated risk of falling carried the phones for 2070 days in which the model classified 14,904,000 events. The system detected 27 of the 37 falls that occurred (sensitivity = 73.0 %) and resulted in one false alarm every 46 days (specificity > 99.9 %, precision = 37.5 %). 42.2 % of the events falsely classified as falls were validated as stumbles.

CONCLUSIONS

The system's performance shows the potential of using smartphones for fall detection and notification in real-life. Apart from functioning as a practical fall monitoring instrument, this system may serve as a valuable research tool, enable future studies to scale their ability to capture fall-related data, and help researchers and clinicians to investigate real-falls.

NT-FDS-A Noise Tolerant Fall Detection System Using Deep Learning on Wearable Devices

Given the high prevalence and detrimental effects of unintentional falls in the elderly, fall detection has become a pertinent public concern. A Fall Detection System (FDS) gathers information from sensors to distinguish falls from routine activities in order to provide immediate medical assistance. Hence, the integrity of collected data becomes imperative. Presence of missing values in data, caused by unreliable data delivery, lossy sensors, local interference and synchronization disturbances and so forth, greatly hamper the credibility and usefulness of data making it unfit for reliable fall detection. This paper presents a noise tolerant FDS performing in presence of missing values in data. The work focuses on Deep Learning (DL) particularly Recurrent Neural Networks (RNNs) with an underlying Bidirectional Long Short-Term Memory (BiLSTM) stack to implement FDS based on wearable sensors. The proposed technique is evaluated on two publicly available datasets—SisFall and UP-Fall Detection. Our system produces an accuracy of 97.21% and 97.41%, sensitivity of 96.97% and 99.77% and specificity of 93.18% and 91.45% on SisFall and UP-Fall Detection respectively, thus outperforming the existing state of the art on these benchmark datasets. The resultant outcomes suggest that the ability of BiLSTM to retain long term dependencies from past and future make it an appropriate model choice to handle missing values for wearable fall detection systems.

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.

Usefulness of Mobile Devices in the Diagnosis and Rehabilitation of Patients with Dizziness and Balance Disorders: A State of the Art Review

Objective

The gold standard for objective body posture examination is posturography. Body movements are detected through the use of force platforms that assess static and dynamic balance (conventional posturography). In recent years, new technologies like wearable sensors (mobile posturography) have been applied during complex dynamic activities to diagnose and rehabilitate balance disorders. They are used in healthy people, especially in the aging population, for detecting falls in the older adults, in the rehabilitation of different neurological, osteoarticular, and muscular system diseases, and in vestibular disorders. Mobile devices are portable, lightweight, and less expensive than conventional posturography. The vibrotactile system can consist of an accelerometer (linear acceleration measurement), gyroscopes (angular acceleration measurement), and magnetometers (heading measurement, relative to the Earth’s magnetic field). The sensors may be mounted to the trunk (most often in the lumbar region of the spine, and the pelvis), wrists, arms, sternum, feet, or shins. Some static and dynamic clinical tests have been performed with the use of wearable sensors. Smartphones are widely used as a mobile computing platform and to evaluate the results or monitor the patient during the movement and rehabilitation. There are various mobile applications for smartphone-based balance systems. Future research should focus on validating the sensitivity and reliability of mobile device measurements compared to conventional posturography.

Conclusion

Smartphone based mobile devices are limited to one sensor lumbar level posturography and offer basic clinical evaluation. Single or multi sensor mobile posturography is available from different manufacturers and offers single to multi-level measurements, providing more data and in some instances even performing sophisticated clinical balance tests.

Comparative Analysis of Real-Time Fall Detection Using Fuzzy Logic Web Services and Machine Learning

Falls are the main cause of susceptibility to severe injuries in many humans, especially for older adults aged 65 and over. Typically, falls are being unnoticed and interpreted as a mere inevitable accident. Various wearable fall warning devices have been created recently for older people. However, most of these devices are dependent on local data processing. Various algorithms are used in wearable sensors to track a real-time fall effectively, which focuses on fall detection via fuzzy-as-a-service based on IEEE 1855–2016, Java Fuzzy Markup Language (FML) and service-oriented architecture. Moreover, several approaches are used to detect a fall using machine learning techniques via human movement positional data to avert any accidents. For fuzzy logic web services, analysis is performed using wearable accelerometer and gyroscope sensors, whereas in machine learning techniques, k-NN, decision tree, random forest and extreme gradient boost are used to differentiate between a fall and non-fall. This study aims to carry out a comparative analysis of real-time fall detection using fuzzy logic web services and machine learning techniques and aims to determine which one is better for real-time fall detection. Research findings exhibit that the proposed fuzzy-as-a-service could easily differentiate between fall and non-fall occurrences in a real-time environment with an accuracy, sensitivity and specificity of 90%, 88.89% and 91.67%, respectively, while the random forest algorithm of machine learning achieved 99.19%, 98.53% and 99.63%, respectively.

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.

Detection of Pre-Impact Falls from Heights Using an Inertial Measurement Unit Sensor

Many safety accidents can occur in industrial sites. Among them, falls from heights (FFHs) are the most frequent accidents and have the highest fatality rate. Therefore, some existing studies have developed personal wearable airbags to mitigate the damage caused by FFHs. To utilize these airbags effectively, it is essential to detect FFHs before collision with the floor. In this study, an inertial measurement unit (IMU) sensor attached to the seventh thoracic vertebrae (T7) was used to develop an FFH detection algorithm. The vertical angle and vertical velocity were calculated using the inertial data obtained from the IMU sensor. Forty young and healthy males were recruited to perform non-FFH and FFH motions. In addition, experiments using a human mannequin and dynamics simulations were performed to obtain FFH data at heights above 2 m. The developed algorithm achieved 100% FFH detection accuracy and provided sufficient lead time such that the airbags could be inflated completely before collision with the floor.

Quantification of Postures for Low-Height Object Manipulation Conducted by Manual Material Handlers in a Retail Environment

Occupational Applications Manual material handlers performing stocking tasks spent substantial amounts of time in bent postures but used traditional stoops and squats infrequently. Instead, they often used split-legged stoops and squats, where one foot is further forward than the other, and one-legged ("golfer's") lifts. During object manipulation, the distance workers reached away from their body, and the height at which they manipulated objects, were correlated with the posture used by the worker. Workers also stayed in different postures for different lengths of time. It is likely that certain postures are more comfortable for the workers to remain in, provide additional mobility or operational radius, or require less energy to use. Understanding these factors in more detail could lead to improved worker training programs, where the postures taught not only have low injury risk but are comfortable so are actually adopted and used by the workers.

Wearable Stretch Sensors for Human Movement Monitoring and Fall Detection in Ergonomics

Wearable sensors are beneficial for continuous health monitoring, movement analysis, rehabilitation, evaluation of human performance, and for fall detection. Wearable stretch sensors are increasingly being used for human movement monitoring. Additionally, falls are one of the leading causes of both fatal and nonfatal injuries in the workplace. The use of wearable technology in the workplace could be a successful solution for human movement monitoring and fall detection, especially for high fall-risk occupations. This paper provides an in-depth review of different wearable stretch sensors and summarizes the need for wearable technology in the field of ergonomics and the current wearable devices used for fall detection. Additionally, the paper proposes the use of soft-robotic-stretch (SRS) sensors for human movement monitoring and fall detection. This paper also recapitulates the findings of a series of five published manuscripts from ongoing research that are published as Parts I to V of “Closing the Wearable Gap” journal articles that discuss the design and development of a foot and ankle wearable device using SRS sensors that can be used for fall detection. The use of SRS sensors in fall detection, its current limitations, and challenges for adoption in human factors and ergonomics are also discussed.

A Study of the Use of Gyroscope Measurements in Wearable Fall Detection Systems

Due to the serious impact of falls on the quality of life of the elderly and on the economical sustainability of health systems, the study of new monitoring systems capable of automatically alerting about falls has gained much research interest during the last decade. In the field of Human Activity Recognition, Fall Detection Systems (FDSs) can be contemplated as pattern recognition architectures able to discriminate falls from ordinary Activities of Daily Living (ADLs). In this regard, the combined application of cellular communications and wearable devices that integrate inertial sensors offers a cost-efficient solution to track the user mobility almost ubiquitously. Inertial Measurement Units (IMUs) typically utilized for these architectures, embed an accelerometer and a gyroscope. This paper investigates if the use of the angular velocity (captured by the gyroscope) as an input feature of the movement classifier introduces any benefit with respect to the most common case in which the classification decision is uniquely based on the accelerometry signals. For this purpose, the work assesses the performance of a deep learning architecture (a convolutional neural network) which is optimized to differentiate falls from ADLs as a function of the raw data measured by the two inertial sensors (gyroscope and accelerometer). The system is evaluated against on a well-known public dataset with a high number of mobility traces (falls and ADL) measured from the movements of a wide group of experimental users.

Enhanced Algorithm for the Detection of Preimpact Fall for Wearable Airbags

Fall-related injury is a common cause of mortality among the elderly. Hip fractures are especially dangerous and can even be fatal. In this study, a threshold-based preimpact fall detection algorithm was developed for wearable airbags that minimize the impact of falls on the user's body. Acceleration sum vector magnitude (SVM), angular velocity SVM, and vertical angle, calculated using inertial data captured from an inertial measurement unit were used to develop the algorithm. To calculate the vertical angle accurately, a complementary filter with a proportional integral controller was used to minimize integration errors and the effect of external impacts. In total, 30 healthy young men were recruited to simulate 6 types of falls and 14 activities of daily life. The developed algorithm achieved 100% sensitivity, 97.54% specificity, 98.33% accuracy, and an average lead time (i.e., the time between the fall detection and the collision) of 280.25 ± 10.29 ms with our experimental data, whereas it achieved 96.1% sensitivity, 90.5% specificity, and 92.4% accuracy with the SisFall public dataset. This paper demonstrates that the algorithm achieved a high accuracy using our experimental data, which included some highly dynamic motions that had not been tested previously.

Detection of Near Falls Using Wearable Devices: A Systematic Review

BACKGROUND AND PURPOSE

Falls among older people are a serious health issue. Remote detection of near falls may provide a new way to identify older people at high risk of falling. This could enable exercise and fall prevention programs to target the types of near falls experienced and the situations that cause near falls before fall-related injuries occur. The purpose of this systematic review was to summarize and critically examine the evidence regarding the detection of near falls (slips, trips, stumbles, missteps, incorrect weight transfer, or temporary loss of balance) using wearable devices.

METHODS

CINAHL, EMBASE, MEDLINE, Compendex, and Inspec were searched to obtain studies that used a wearable device to detect near falls in young and older people with or without a chronic disease and were published in English.

RESULTS

Nine studies met the final inclusion criteria. Wearable sensors used included accelerometers, gyroscopes, and insole force inducers. The waist was the most common location to place a single device. Both high sensitivity (≥85.7%) and specificity (≥90.0%) were reported for near-fall detection during various clinical simulations and improved when multiple devices were worn. Several methodological issues that increased the risk of bias were revealed. Most studies analyzed a single or few near-fall types by younger adults in controlled laboratory environments and did not attempt to distinguish naturally occurring near falls from actual falls or other activities of daily living in older people.

CONCLUSIONS

The use of a single lightweight sensor to distinguish between different types of near falls, actual falls, and activities of daily living is a promising low-cost technology and clinical tool for long-term continuous monitoring of older people and clinical populations at risk of falls. However, currently the evidence is limited because studies have largely involved simulated laboratory events in young adults. Future studies should focus on validating near-fall detection in larger cohorts and include data from (i) people at high risk of falling, (ii) activities of daily living, (iii) both near falls and actual falls, and (iv) naturally occurring near falls.

Consumption Analysis of Smartphone based Fall Detection Systems with Multiple External Wireless Sensors

Fall Detection Systems (FDSs) based on wearable technologies have gained much research attention in recent years. Due to the networking and computing capabilities of smartphones, these widespread personal devices have been proposed to deploy cost-effective wearable systems intended for automatic fall detection. In spite of the fact that smartphones are natively provided with inertial sensors (accelerometers and gyroscopes), the effectiveness of a smartphone-based FDS can be improved if it also exploits the measurements collected by small low-power wireless sensors, which can be firmly attached to the user’s body without causing discomfort. For these architectures with multiple sensing points, the smartphone transported by the user can act as the core of the FDS architecture by processing and analyzing the data measured by the external sensors and transmitting the corresponding alarm whenever a fall is detected. In this context, the wireless communications with the sensors and with the remote monitoring point may impact on the general performance of the smartphone and, in particular, on the battery lifetime. In contrast with most works in the literature (which disregard the real feasibility of implementing an FDS on a smartphone), this paper explores the actual potential of current commercial smartphones to put into operation an FDS that incorporates several external sensors. This study analyzes diverse operational aspects that may influence the consumption (as the use of a GPS sensor, the coexistence with other apps, the retransmission of the measurements to an external server, etc.) and identifies practical scenarios in which the deployment of a smartphone-based FDS is viable.

Pre-Impact Fall Detection Using 3D Convolutional Neural Network

Early fall detection is an important issue during gait rehabilitation training. This paper proposes an approach for pre-impact fall detection during gait rehabilitation training based on a 3D convolutional neural network (CNN). Firstly, pre-training data is collected and used to pre-train the 3D CNN to differentiate between a normal walking and a fall based on their general spatio-temporal patterns. Secondly, fine-tuning data is created by combining the pre-training data with a 3second normal walking sample collected from a new trainee whose falls are to be detected. The pre-trained 3D CNN is further fine-tuned by the fine-tuning data to learn the spatiotemporal patterns of the new trainee. Finally, a temporal sliding window is used to feed video snippets into the fine-tuned 3D CNN for fall detection. To the best of our knowledge, this is the first pre-impact fall detection approach based on a 3D CNN using RGB images. Moreover, the training strategy used to train the 3D CNN can alleviate the generalization issue of the 3D CNN when only limited training data is available in gait rehabilitation training. With 225 testing trials from 5 trainees, the proposed pre-impact fall detection approach achieves a detection accuracy of 100% within 0.5 second after falls start. Experiment results show that this approach is efficient, accurate, and practical in achieving intelligent fall detection during gait rehabilitation training.

Evaluation of time-frequency features as detectors of lack of balance due to tripping-like perturbations

Unbalancing events during gait can end up in falls and, thus, injury. Detecting events that could bring to fall and consequently activating fall prevention systems before the impact may help to mitigate related injuries. However, there is uncertainty about signals and methods that could offer the best performance. In this paper we investigated a novel trip detection method based on time-frequency features to evaluate the performances of these features as trip detectors. Hip angles of eight healthy young subjects were recorded while performing unexpected tripping trials delivered during steady locomotion. Then the Short-Time Fourier Transform (STFT) of the hip angle was estimated. Median frequency, power, centroidal frequency as well as frequency dispersion were computed for each time sliced power spectrum. These features were used as input for a trip detection algorithm. We assessed detection time (T_detect), specificity (Spec) and sensitivity (Sens) for each feature. Performances obtained with median frequencies over time(T_detect 0.91 ± 0.47 s; Sens 0.96) were better than those obtained using the hip angle signal in time domain (T_detect 1.19 ± 0.27 s; Sens 0.83). Other features did not show significant results. Thus, median frequency over time expected to achieve effective real-time event detection systems, with the aim of a future on-board application concerning detection and prevention measures.

A Novel Sensor Data Pre-Processing Methodology for the Internet of Things Using Anomaly Detection and Transfer-By-Subspace-Similarity Transformation

The Internet of Things (IoT) and sensors are becoming increasingly popular, especially in monitoring large and ambient environments. Applications that embrace IoT and sensors often require mining the data feeds that are collected at frequent intervals for intelligence. Despite the fact that such sensor data are massive, most of the data contents are identical and repetitive; for example, human traffic in a park at night. Most of the traditional classification algorithms were originally formulated decades ago, and they were not designed to handle such sensor data effectively. Hence, the performance of the learned model is often poor because of the small granularity in classification and the sporadic patterns in the data. To improve the quality of data mining from the IoT data, a new pre-processing methodology based on subspace similarity detection is proposed. Our method can be well integrated with traditional data mining algorithms and anomaly detection methods. The pre-processing method is flexible for handling similar kinds of sensor data that are sporadic in nature that exist in many ambient sensing applications. The proposed methodology is evaluated by extensive experiment with a collection of classical data mining models. An improvement over the precision rate is shown by using the proposed method.

Closing the Wearable Gap—Part III: Use of Stretch Sensors in Detecting Ankle Joint Kinematics During Unexpected and Expected Slip and Trip Perturbations

Background: An induced loss of balance resulting from a postural perturbation has been reported as the primary source for postural instability leading to falls. Hence; early detection of postural instability with novel wearable sensor-based measures may aid in reducing falls and fall-related injuries. The purpose of the study was to validate the use of a stretchable soft robotic sensor (SRS) to detect ankle joint kinematics during both unexpected and expected slip and trip perturbations. Methods: Ten participants (age: 23.7 ± 3.13 years; height: 170.47 ± 8.21 cm; mass: 82.86 ± 23.4 kg) experienced a counterbalanced exposure of an unexpected slip, an unexpected trip, an expected slip, and an expected trip using treadmill perturbations. Ankle joint kinematics for dorsiflexion and plantarflexion were quantified using three-dimensional (3D) motion capture through changes in ankle joint range of motion and using the SRS through changes in capacitance when stretched due to ankle movements during the perturbations. Results: A greater R-squared and lower root mean square error in the linear regression model was observed in comparing ankle joint kinematics data from motion capture with stretch sensors. Conclusions: Results from the study demonstrated that 71.25% of the trials exhibited a minimal error of less than 4.0 degrees difference from the motion capture system and a greater than 0.60 R-squared value in the linear model; suggesting a moderate to high accuracy and minimal errors in comparing SRS to a motion capture system. Findings indicate that the stretch sensors could be a feasible option in detecting ankle joint kinematics during slips and trips.

A Novel Activity Recognition System for Alternative Control Strategies of a Lower Limb Rehabilitation Robot

Robot-aided training strategies that allow functional, assist-as-needed, or challenging training have been widely explored. Accurate activity recognition is the basis for implementing alternative training strategies. However, some obstacles to accurate recognition exist. First, scientists do not yet fully understand some rehabilitation activities, such as abnormal gaits and falls; thus, there is no standardized feature for identifying such activities. Second, during the activity identification process, it is difficult to reasonably balance sensitivity and specificity when setting the threshold. Therefore, we proposed a multisensor fusion system and a two-stage activity recognition classifier. This multisensor system integrates explicit information such as kinematics and spatial distribution information along with implicit information such as kinetics and pulse information. Both the explicit and implicit information are analyzed in one discriminant function to obtain a detailed and accurate recognition result. Then, alternative training strategies can be implemented on this basis. Finally, we conducted experiments to verify the feasibility and efficiency of the multisensor fusion system. The experimental results show that the proposed fusion system achieves an accuracy of 99.37%, and the time required to prejudge a fall is approximately 205 milliseconds faster than the response time of single-sensor systems. Moreover, the proposed system also identifies fall directions and abnormal gait types.

Pre-Impact Detection Algorithm to Identify Tripping Events Using Wearable Sensors

This study aimed to investigate the performance of an updated version of our pre-impact detection algorithm parsing out the output of a set of Inertial Measurement Units (IMUs) placed on lower limbs and designed to recognize signs of lack of balance due to tripping. Eight young subjects were asked to manage tripping events while walking on a treadmill. An adaptive threshold-based algorithm, relying on a pool of adaptive oscillators, was tuned to identify abrupt kinematics modifications during tripping. Inputs of the algorithm were the elevation angles of lower limb segments, as estimated by IMUs located on thighs, shanks and feet. The results showed that the proposed algorithm can identify a lack of balance in about 0.37 ± 0.11 s after the onset of the perturbation, with a low percentage of false alarms (<10%), by using only data related to the perturbed shank. The proposed algorithm can hence be considered a multi-purpose tool to identify different perturbations (i.e., slippage and tripping). In this respect, it can be implemented for different wearable applications (e.g., smart garments or wearable robots) and adopted during daily life activities to enable on-demand injury prevention systems prior to fall impacts.

Wearable Sensors System for an Improved Analysis of Freezing of Gait in Parkinson's Disease Using Electromyography and Inertial Signals

We propose a wearable sensor system for automatic, continuous and ubiquitous analysis of Freezing of Gait (FOG), in patients affected by Parkinson’s disease. FOG is an unpredictable gait disorder with different clinical manifestations, as the trembling and the shuffling-like phenotypes, whose underlying pathophysiology is not fully understood yet. Typical trembling-like subtype features are lack of postural adaptation and abrupt trunk inclination, which in general can increase the fall probability. The targets of this work are detecting the FOG episodes, distinguishing the phenotype and analyzing the muscle activity during and outside FOG, toward a deeper insight in the disorder pathophysiology and the assessment of the fall risk associated to the FOG subtype. To this aim, gyroscopes and surface electromyography integrated in wearable devices sense simultaneously movements and action potentials of antagonist leg muscles. Dedicated algorithms allow the timely detection of the FOG episode and, for the first time, the automatic distinction of the FOG phenotypes, which can enable associating a fall risk to the subtype. Thanks to the possibility of detecting muscles contractions and stretching exactly during FOG, a deeper insight into the pathophysiological underpinnings of the different phenotypes can be achieved, which is an innovative approach with respect to the state of art.

Evaluation of Inertial Sensor-Based Pre-Impact Fall Detection Algorithms Using Public Dataset

In this study, pre-impact fall detection algorithms were developed based on data gathered by a custom-made inertial measurement unit (IMU). Four types of simulated falls were performed by 40 healthy subjects (age: 23.4 ± 4.4 years). The IMU recorded acceleration and angular velocity during all activities. Acceleration, angular velocity, and trunk inclination thresholds were set to 0.9 g, 47.3°/s, and 24.7°, respectively, for a pre-impact fall detection algorithm using vertical angles (VA algorithm); and 0.9 g, 47.3°/s, and 0.19, respectively, for an algorithm using the triangle feature (TF algorithm). The algorithms were validated by the results of a blind test using four types of simulated falls and six types of activities of daily living (ADL). VA and TF algorithms resulted in lead times of 401 ± 46.9 ms and 427 ± 45.9 ms, respectively. Both algorithms were able to detect falls with 100% accuracy. The performance of the algorithms was evaluated using a public dataset. Both algorithms detected every fall in the SisFall dataset with 100% sensitivity). The VA algorithm had a specificity of 78.3%, and TF algorithm had a specificity of 83.9%. The algorithms had higher specificity when interpreting data from elderly subjects. This study showed that algorithms using angles could more accurately detect falls. Public datasets are needed to improve the accuracy of the algorithms.

Kinect4FOG: monitoring and improving mobility in people with Parkinson's using a novel system incorporating the Microsoft Kinect v2

Parkinson's is a neurodegenerative condition associated with several motor symptoms including tremors and slowness of movement. Freezing of gait (FOG); the sensation of one's feet being "glued" to the floor, is one of the most debilitating symptoms associated with advanced Parkinson's. FOG not only contributes to falls and related injuries, but also compromises quality of life as people often avoid engaging in functional daily activities both inside and outside the home. In the current study, we describe a novel system designed to detect FOG and falling in people with Parkinson's (PwP) as well as monitoring and improving their mobility using laser-based visual cues cast by an automated laser system. The system utilizes a RGB-D sensor based on Microsoft Kinect v2 and a laser casting system consisting of two servo motors and an Arduino microcontroller. This system was evaluated by 15 PwP with FOG. Here, we present details of the system along with a summary of feedback provided by PwP. Despite limitations regarding its outdoor use, feedback was very positive in terms of domestic usability and convenience, where 12/15 PwP showed interest in installing and using the system at their homes. Implications for Rehabilitation Providing an automatic and remotely manageable monitoring system for PwP gait analysis and fall detection. Providing an automatic, unobtrusive and dynamic visual cue system for PwP based on laser line projection. Gathering feedback from PwP about the practical usage of the implemented system through focus group events.

On the Comparison of Wearable Sensor Data Fusion to a Single Sensor Machine Learning Technique in Fall Detection

In the context of the ageing global population, researchers and scientists have tried to find solutions to many challenges faced by older people. Falls, the leading cause of injury among elderly, are usually severe enough to require immediate medical attention; thus, their detection is of primary importance. To this effect, many fall detection systems that utilize wearable and ambient sensors have been proposed. In this study, we compare three newly proposed data fusion schemes that have been applied in human activity recognition and fall detection. Furthermore, these algorithms are compared to our recent work regarding fall detection in which only one type of sensor is used. The results show that fusion algorithms differ in their performance, whereas a machine learning strategy should be preferred. In conclusion, the methods presented and the comparison of their performance provide useful insights into the problem of fall detection.

Detection of American Football Head Impacts Using Biomechanical Features and Support Vector Machine Classification

Accumulation of head impacts may contribute to acute and long-term brain trauma. Wearable sensors can measure impact exposure, yet current sensors do not have validated impact detection methods for accurate exposure monitoring. Here we demonstrate a head impact detection method that can be implemented on a wearable sensor for detecting field football head impacts. Our method incorporates a support vector machine classifier that uses biomechanical features from the time domain and frequency domain, as well as model predictions of head-neck motions. The classifier was trained and validated using instrumented mouthguard data from collegiate football games and practices, with ground truth data labels established from video review. We found that low frequency power spectral density and wavelet transform features (10~30 Hz) were the best performing features. From forward feature selection, fewer than ten features optimized classifier performance, achieving 87.2% sensitivity and 93.2% precision in cross-validation on the collegiate dataset (n = 387), and over 90% sensitivity and precision on an independent youth dataset (n = 32). Accurate head impact detection is essential for studying and monitoring head impact exposure on the field, and the approach in the current paper may help to improve impact detection performance on wearable sensors.

A data-driven approach to modeling physical fatigue in the workplace using wearable sensors

Wearable sensors are currently being used to manage fatigue in professional athletics, transportation and mining industries. In manufacturing, physical fatigue is a challenging ergonomic/safety "issue" since it lowers productivity and increases the incidence of accidents. Therefore, physical fatigue must be managed. There are two main goals for this study. First, we examine the use of wearable sensors to detect physical fatigue occurrence in simulated manufacturing tasks. The second goal is to estimate the physical fatigue level over time. In order to achieve these goals, sensory data were recorded for eight healthy participants. Penalized logistic and multiple linear regression models were used for physical fatigue detection and level estimation, respectively. Important features from the five sensors locations were selected using Least Absolute Shrinkage and Selection Operator (LASSO), a popular variable selection methodology. The results show that the LASSO model performed well for both physical fatigue detection and modeling. The modeling approach is not participant and/or workload regime specific and thus can be adopted for other applications.

A wearable action recognition system based on acceleration and attitude angles using real-time detection algorithm

Falls are a main cause of trauma and death. The purpose of this study is to adopt unique resultant acceleration and attitude angles to distinguish falls from activities of daily life before impact. In this study, we developed a wearable action recognition system to acquire action data. The moving average filter was employed to deal with raw data, and then complementary filter was adopted to compromise sensor data for attitude angles. The real-time detection algorithm embedded in this device was applied to recognize six actions based on processed data. Eight subjects (five males, three females) participated in the experiment. The optimal features and related thresholds were extracted. In addition, the real-time action detection results indicated that the real-time action recognition model reached an accuracy of 96.25%, with 98% for male and 93.3% for female. Thus, our device potentially achieves a high sensitivity of fall-related actions recognition.

Falls are unintentional: Studying simulations is a waste of faking time

Researchers tend to agree that falls are, by definition, unintentional and that sensor algorithms (the processes that allows a computer program to identify a fall among data from sensors) perform poorly when attempting to detect falls 'in the wild' (a phrase some scientists use to mean 'in reality'). Algorithm development has been reliant on simulation, i.e. asking actors to throw themselves intentionally to the ground. This is unusual (no one studies faked coughs or headaches) and uninformative (no one can intend the unintentional). Researchers would increase their chances of detecting 'real' falls in 'the real world' by studying the behaviour of fallers, however, vulnerable, before, during and after the event: the literature on the circumstances of falling is more informative than any number of faked approximations. A complimentary knowledge base (in falls, sensors and/or signals) enables multidisciplinary teams of clinicians, engineers and computer scientists to tackle fall detection and aim for fall prevention. Throughout this paper, I discuss differences between falls, 'intentional falling' and simulations, and the balance between simulation and reality in falls research, finally suggesting ways in which researchers can access examples of falls without resorting to fakery.

Prediction of Intrinsically Caused Tripping Events in Individuals With Stroke

This study investigated the feasibility of predicting intrinsically caused trips (ICTs) in individuals with stroke. Gait kinematics collected from 12 individuals with chronic stroke, who demonstrated ICTs in treadmill walking, were analyzed. A prediction algorithm based on the outlier principle was employed. Sequential forward selection (SFS) and minimum-redundancy-maximum-relevance (mRMR) were used separately to identify the precursors for accurate ICT prediction. The results showed that it was feasible to predict ICTs around 50-260 ms before ICTs occurred in the swing phase by monitoring lower limb kinematics during the preceding stance phase. Both SFS and mRMR were effective in identifying the precursors of ICTs. For 9 out of the 12 subjects, the paretic lower limb's shank orientation in the sagittal plane and the vertical velocity of the paretic foot's center of gravity were important in predicting ICTs accurately; the averaged area under receiver operating characteristic curve achieved 0.95 and above. For the other three subjects, kinematics of the less affected limb or proximal joints in the paretic side were identified as the precursors to an ICT, potentially due to the variations of neuromotor deficits among stroke survivors. Although additional engineering efforts are still needed to address the challenges in making our design clinically practical, the outcome of this study may lead to further proactive engineering mechanisms for ICT avoidance and therefore reduce the risk of falls in individuals with stroke.

Optimal fall indicators for slip induced falls on a cross-slope

Slip-induced falls are among the most common cause of major occupational injuries in the UK as well as being a major public health concern in the elderly population. This study aimed to determine the optimal fall indicators for fall detection models which could be used to reduce the detrimental consequences of falls. A total of 264 kinematic variables covering three-dimensional full body model translation and rotational measures were analysed during normal walking, successful recovery from slips and falls on a cross-slope. Large effect sizes were found for three kinematic variables which were able to distinguish falls from normal walking and successful recovery. Further work should consider other types of daily living activities as results show that the optimal kinematic fall indicators can vary considerably between movement types. Practitioner Summary: Fall detection models are used to minimise the adverse consequences of slip-induced falls, a major public health concern. Optimal fall indicators were derived from a comprehensive set of kinematic variables for slips on a cross-slope. Results suggest robust detection of falls is possible on a cross-slope but may be more difficult than level walking.

Pre-impact fall detection

Pre-impact fall detection has been proposed to be an effective fall prevention strategy. In particular, it can help activate on-demand fall injury prevention systems (e.g. inflatable hip protectors) prior to fall impacts, and thus directly prevent the fall-related physical injuries. This paper gave a systematical review on pre-impact fall detection, and focused on the following aspects of the existing pre-impact fall detection research: fall detection apparatus, fall detection indicators, fall detection algorithms, and types of falls for fall detection evaluation. In addition, the performance of the existing pre-impact fall detection solutions were also reviewed and reported in terms of their sensitivity, specificity, and detection/lead time. This review also summarized the limitations in the existing pre-impact fall detection research, and proposed future research directions in this field.

The Use of Empirical Mode Decomposition-Based Algorithm and Inertial Measurement Units to Auto-Detect Daily Living Activities of Healthy Adults

he use of inertial measurement units (IMUs) in motion analysis for clinical purpose is relatively recent. However, the use of such system in free environment remains sparse. This is in part due the lack of robust algorithms to handle large volumes of data for performance evaluation and patient diagnosis. The present work examines the ability of using Empirical Mode Decomposition and discrete-time detection of events to automatically detect and segment tasks associated with activities of daily living (ADL) using IMUs. Seven healthy older adults (73± 4 years old) performed ADL tasks in a simulated apartment during trials of different durations (3, 4, and 5-min). They wore a suit (Synertial-IGS180) comprised of 17-IMUs positioned strategically on body segments to capture full body motion. After a systematic process examining time series of each sensor, it was determined that 6-IMUs were sufficient to detect the 9 tasks at hand (such as walking, sit to stand, stand to sit, reaching to the ground to pick or to put down objects on the floor, step an obstacle and turning). The proposed method automatically identified the proper set of template waveforms associated to ADL tasks based on kinematic data acquired from the selected IMUs. The ground truth on timing of tasks was established by visual segmentation of recordings using the system's software. Despite the variation in the occurrences of the performed tasks (freely moving), the proposed algorithm exhibited high global accuracy under unscripted conditions of motion, for both Se. and Sp. of 97% (Nevents=1999), using a few features and without learning process. This work will eventually allow for the assessment of mobility performance within the segmented signals; specifically how well the person is moving in his/her environment.

Forward kinematics using IMU on-body sensor network for mobile analysis of human kinematics

The feasibility of large network inertial measurement units (IMUs) are evaluated for purposes requiring feedback. A series of wireless IMUs were attached to a human lower-limb laboratory model outfitted with joint angle encoders. The goal was to discover if large networks of wireless IMUs can give realtime joint orientation data while still maintaining an acceptable degree of accuracy.

Prior-to- and Post-Impact Fall Detection Using Inertial and Barometric Altimeter Measurements

This paper investigates a fall detection system based on the integration of an inertial measurement unit with a barometric altimeter (BIMU). The vertical motion of the body part the BIMU was attached to was monitored on-line using a method that delivered drift-free estimates of the vertical velocity and estimates of the height change from the floor. The experimental study included activities of daily living of seven types and falls of five types, simulated by a cohort of 25 young healthy adults. The downward vertical velocity was thresholded at 1.38 m/s, yielding 80% sensitivity (SE), 100% specificity (SP) and a mean prior-to-impact time of 157 ms (range 40-300 ms). The soft falls, i.e., those with downward vertical velocity above 0.55 m/s and below 1.38 m/s were analyzed post-impact. Six fall detection methods, tuned to achieve 100% SE, were considered to include features of impact, change of posture and height, singularly or in association with one another. No single feature allowed for 100% SP. The detection accuracy marginally improved when the height change was considered in association with either the impact or the change of posture; the post-impact fall detection method that analyzed the impact and the change of posture together achieved 100% SP.