Selecting Power-Efficient Signal Features for a Low-Power Fall Detector

Technologies for monitoring activities of daily living in older adults: a systematic review

PURPOSE

As the older adult population rise globally, technologies to monitoring activities of daily living (ADL) may have a role in supporting aging in place for older adults. The objective of this systematic literature review was to study the scope, diversity and readiness of technologies developed to monitor ADL in older adults.

METHODS

We systematically searched two scientific databases (CINAHL and IEEE), following Preferred Reporting Items for Systematic reviews and Meta Analyses (PRISMA) guidelines. We included studies on technologies used to monitor older adults' ADL in the home but excluded studies focused on communication technologies (phone calls, text messages) or monitoring postures alone. The JBI checklist for case series was used for quality assessment. Extracted details included population characteristics, ADL assessment outcomes, types of monitoring technology, and technology readiness and usability.

RESULTS

The search found 147 papers, with 16 papers included in the final analysis. The literature described 48 types of technologies. Of moderate quality studies, five studies used wearables at technology readiness level 4-6 to monitor basic ADL (walking, transfers and walking up stairs) and one used ambient sensors to detect urinary incontinence.

CONCLUSIONS

Monitoring technologies remain at development stages. More research is needed to strengthen technologies that monitor activities of daily living.

An Unobtrusive Fall Detection System Using Low Resolution Thermal Sensors and Convolutional Neural Networks

Human activity recognition has many potential applications. In an aged care facility, it is crucial to monitor elderly patients and assist them in the case of falls or other needs. Wearable devices can be used for such a purpose. However, most of them have been proven to be obtrusive, and patients reluctate or forget to wear them. In this study, we used infrared technology to recognize certain human activities including sitting, standing, walking, laying in bed, laying down, and falling. We evaluated a system consisting of two 24×32 thermal array sensors. One infrared sensor was installed on side and another one was installed on the ceiling of an experimental room capturing the same scene. We chose side and overhead mounts to compare the performance of classifiers. We used our prototypes to collect data from healthy young volunteers while performing eight different scenarios. After that, we converted data coming from the sensors into images and applied a supervised deep learning approach. The scene was captured by a visible camera and the video from the visible camera was used as the ground truth. The deep learning network consisted of a convolutional neural network which automatically extracted features from infrared images. Overall average F1-score of all classes for the side mount was 0.9044 and for the overhead mount was 0.8893. Overall average accuracy of all classes for the side mount was 96.65% and for the overhead mount was 95.77%. Our results suggested that our infrared-based method not only could unobtrusively recognize human activities but also was reasonably accurate.

Domain-Adaptive Fall Detection Using Deep Adversarial Training

Fall detection (FD) systems are important assistive technologies for healthcare that can detect emergency fall events and alert caregivers. However, it is not easy to obtain large-scale annotated fall events with various specifications of sensors or sensor positions during the implementation of accurate FD systems. Moreover, the knowledge obtained through machine learning has been restricted to tasks in the same domain. The mismatch between different domains might hinder the performance of FD systems. Cross-domain knowledge transfer is very beneficial for machine-learning based FD systems to train a reliable FD model with well-labeled data in new environments. In this study, we propose domain-adaptive fall detection (DAFD) using deep adversarial training (DAT) to tackle cross-domain problems, such as cross-position and cross-configuration. The proposed DAFD can transfer knowledge from the source domain to the target domain by minimizing the domain discrepancy to avoid mismatch problems. The experimental results show that the average F1-score improvement when using DAFD ranges from 1.5% to 7% in the cross-position scenario, and from 3.5% to 12% in the cross-configuration scenario, compared to using the conventional FD model without domain adaptation training. The results demonstrate that the proposed DAFD successfully helps to deal with cross-domain problems and to achieve better detection performance.

On the Heterogeneity of Existing Repositories of Movements Intended for the Evaluation of Fall Detection Systems

Due to the serious impact of falls on the autonomy and health of older people, the investigation of wearable alerting systems for the automatic detection of falls has gained considerable scientific interest in the field of body telemonitoring with wireless sensors. Because of the difficulties of systematically validating these systems in a real application scenario, Fall Detection Systems (FDSs) are typically evaluated by studying their response to datasets containing inertial sensor measurements captured during the execution of labelled nonfall and fall movements. In this context, during the last decade, numerous publicly accessible databases have been released aiming at offering a common benchmarking tool for the validation of the new proposals on FDSs. This work offers a comparative and updated analysis of these existing repositories. For this purpose, the samples contained in the datasets are characterized by different statistics that model diverse aspects of the mobility of the human body in the time interval where the greatest change in the acceleration module is identified. By using one-way analysis of variance (ANOVA) on the series of these features, the comparison shows the significant differences detected between the datasets, even when comparing activities that require a similar degree of physical effort. This heterogeneity, which may result from the great variability of the sensors, experimental users, and testbeds employed to generate the datasets, is relevant because it casts doubt on the validity of the conclusions of many studies on FDSs, since most of the proposals in the literature are only evaluated using a single database.

Accelerometer-Based Fall Detection Using Machine Learning: Training and Testing on Real-World Falls

Falling is a significant health problem. Fall detection, to alert for medical attention, has been gaining increasing attention. Still, most of the existing studies use falls simulated in a laboratory environment to test the obtained performance. We analyzed the acceleration signals recorded by an inertial sensor on the lower back during 143 real-world falls (the most extensive collection to date) from the FARSEEING repository. Such data were obtained from continuous real-world monitoring of subjects with a moderate-to-high risk of falling. We designed and tested fall detection algorithms using features inspired by a multiphase fall model and a machine learning approach. The obtained results suggest that algorithms can learn effectively from features extracted from a multiphase fall model, consistently overperforming more conventional features. The most promising method (support vector machines and features from the multiphase fall model) obtained a sensitivity higher than 80%, a false alarm rate per hour of 0.56, and an F-measure of 64.6%. The reported results and methodologies represent an advancement of knowledge on real-world fall detection and suggest useful metrics for characterizing fall detection systems for real-world use.

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.

Internet of Things (IoT)-Enabled Elderly Fall Verification, Exploiting Temporal Inference Models in Smart Homes

Everyday life of the elderly and impaired population living in smart homes is challenging because of possible accidents that may occur due to daily activities. In such activities, persons often lean over (to reach something) and, if they not cautious, are prone to falling. To identify fall incidents, which could stochastically cause serious injuries or even death, we propose specific temporal inference models; namely, CM-I and CM-II. These models can infer a fall incident based on classification methods by exploiting wearable Internet of Things (IoT) altimeter sensors adopted by seniors. We analyzed real and synthetic data of fall and lean over incidents to test the proposed models. The results are promising for incorporating such inference models to assist healthcare for fall verification of seniors in smart homes. Specifically, the CM-II model achieved a prediction accuracy of 0.98, which is the highest accuracy when compared to other models in the literature under the McNemar's test criterion. These models could be incorporated in wearable IoT devices to provide early warning and prediction of fall incidents to clinical doctors.