Elderly Fall Detection Systems: A Literature Survey

Multimodal dataset for sensor fusion in fall detection

The necessity for effective automatic fall detection mechanisms in older adults is driven by the growing demographic of elderly individuals who are at substantial health risk from falls, particularly when residing alone. Despite the existence of numerous fall detection systems (FDSs) that utilize machine learning and predictive modeling, accurately distinguishing between everyday activities and genuine falls continues to pose significant challenges, exacerbated by the varied nature of residential settings. Adaptable solutions are essential to cater to the diverse conditions under which falls occur. In this context, sensor fusion emerges as a promising solution, harnessing the unique physical properties of falls. The success of developing effective detection algorithms is dependent on the availability of comprehensive datasets that integrate data from multiple synchronized sensors. Our research introduces a novel multisensor dataset designed to support the creation and evaluation of advanced multisensor fall detection algorithms. This dataset was compiled from simulations of ten different fall types by ten participants, ensuring a wide array of scenarios. Data were collected using four types of sensors: a mobile phone equipped with a single-channel, three-dimensional accelerometer; a far infrared (FIR) thermal camera; an $8×8$ LIDAR; and a 60-64 GHz radar. These sensors were selected for their combined effectiveness in capturing detailed aspects of fall events while mitigating privacy issues linked to visual recordings. Characterization of the dataset was undertaken using two key metrics: the instantaneous norm of the signal and the temporal difference between consecutive frames. This analysis highlights the distinct variations between fall and non-fall events across different sensors and signal characteristics. Through the provision of this dataset, our objective is to facilitate the development of sensor fusion algorithms that surpass the accuracy and reliability of traditional single-sensor FDSs.

A Decade of Progress in Wearable Sensors for Fall Detection (2015-2024): A Network-Based Visualization Review

Over the past decade, wearable sensors for fall detection have gained significant attention due to their potential in improving the safety of elderly users and reducing fall-related injuries. This review employs a network-based visualization approach to analyze research trends, key technologies, and collaborative networks. Using studies from SCI- and SSCI-indexed journals from 2015 to 2024, we analyzed 582 articles and 65 reviews with CiteSpace, revealing a significant rise in research on wearable sensors for fall detection. Additionally, we reviewed various datasets and machine learning techniques, from traditional methods to advanced deep learning frameworks, which demonstrate high accuracies, F1 scores, sensitivities, and specificities in controlled settings. This review provides a comprehensive overview of the progress and emerging trends, offering a foundation for future advancements in wearable fall detection systems.

Enhancing Slip, Trip, and Fall Prevention: Real-World Near-Fall Detection with Advanced Machine Learning Technique

Slips, trips, and falls (STFs) are a major occupational hazard that contributes significantly to workplace injuries and the associated financial costs. The application of traditional fall detection techniques in the real world is limited because they are usually based on simulated falls. By using kinematic data from real near-fall incidents that occurred in physically demanding work environments, this study overcomes this limitation and improves the ecological validity of fall detection algorithms. This study systematically tests several machine-learning architectures for near-fall detection using the Prev-Fall dataset, which consists of high-resolution inertial measurement unit (IMU) data from 110 workers. Convolutional neural networks (CNNs), residual networks (ResNets), convolutional long short-term memory networks (convLSTMs), and InceptionTime models were trained and evaluated over a range of temporal window lengths using a neural architecture search. High-validation F1 scores were achieved by the best-performing models, particularly CNNs and InceptionTime, indicating their effectiveness in near-fall classification. The need for more contextual variables to increase robustness was highlighted by recurrent false positives found in subsequent tests on previously unobserved occupational data, especially during biomechanically demanding activities such as bending and squatting. Nevertheless, our findings suggest the applicability of machine-learning-based STF prevention systems for workplace safety monitoring and, more generally, applications in fall mitigation. To further improve the accuracy and generalizability of the system, future research should investigate multimodal data integration and improved classification techniques.

[Longitudinal Transitions of Fall States Based on a Multi-State Markov Model and Their Associated Risk Factors]

Objective

To investigate the transition intensity and transition probabilities of fall states among middle-aged and older adults in China, and to assess the impact of potential risk factors on falls.

Methods

We utilized in the study data from the China Health and Retirement Longitudinal Study (CHARLS) and employed a multi-state Markov model (MSM) to analyze the transition intensity and probabilities between states of no falls or falls without treatment, falls requiring treatment, and death.

Results

A total of 14722 participants were enrolled, with a mean age of (59.4 years ± 9.7 years), and 47.9% were male. The median follow-up period was 9 years (interquartile range [IQR], 7-9 years). At baseline, 12381 participants (84.1%) reported no falls or falls without treatment, while 2341 (15.9%) reported falls requiring treatment. Participants who experienced falls requiring treatment within one follow-up cycle had a 55.2% probability of not falling again or only falling without treatment in the subsequent two years, a 37.6% probability of continuing to experience falls requiring treatment, and a 7.2% probability of death. The risk of transitioning from a state of no falls or falls without treatment to falls requiring treatment increased by 8.6% for every 5-year increase in age. The risk was 35.1% higher for females compared to males. Rural residents had a 10.1% higher risk. Those who were divorced, separated, widowed, or never married had a 20.7% higher risk. Higher degrees of physical function impairment were associated with an increased risk. Depressive symptoms increased the risk by 31.6%. Having one chronic disease raised the risk by 9.6%, while multimorbidity led to a 28.8% increase in risk.

Conclusion

According to the findings of the study, falls are a dynamic process and emphasis should be given to fall prevention for older adults, individuals with a history of fall-related medical visits, those living alone, those with impaired physical function, and those with depressive symptoms.

Human fall simulation testing method: where we are

Falls pose a significant threat to human health and safety. Accurately assessing the protective effectiveness of fall protection products can significantly reduce the occurrence of fall accidents. This paper systematically reviews the types and risk factors of human falls and then discusses the current research status and future prospects of various test methods for human fall protection. A literature search was conducted in databases such as Web of Science, Google Scholar, and Scopus. This study focuses on experimental methods for human fall testing, simulation model experiments, and finite element simulations, providing an outlook on future development trends. In the discussion of three different fall testing methods, research indicates that human fall simulation testing faces limitations such as ethical concerns and safety issues. Although simulation experiments allow for multiple tests in a short period, the complexity and accuracy of the models may affect the reliability of the results. By integrating more experimental data, optimizing the design of human models, and incorporating finite element simulation technology, the scope of testing applications can be expanded, thereby improving the effectiveness of protective product designs. In conclusion, future research on fall protection testing methods should aim to establish unified international standards, which will enhance consistency and repeatability in testing, facilitating better comparison and evaluation of the effectiveness of various protective measures. Furthermore, the integration of more experimental data with real-world scenarios, the optimization of human models and test environments, and the promotion of finite element simulation technology will be crucial in enhancing the precision of protective assessments.

Fall Risk Prediction Using Instrumented Footwear in Institutionalized Older Adults

This study presents a novel framework that utilizes instrumented footwear to predict fall risk in institutionalized older adults by leveraging stride-to-stride gait data. The older adults are categorized into fallers and non-fallers using three distinct criteria: retrospective fall history, prospective fall occurrence, and a combination of both retrospective and prospective data. Three types of data collected from N=95 institutionalized older adults are analyzed: traditional timed mobility tests, gait data collected from a validated electronic walkway, and gait data collected with instrumented footwear developed by our team. The importance of each type of data is assessed using a brute-force search method, through which the optimal features are selected. AdaBoost algorithms are then utilized to develop predictive models based on the selected features. The models are evaluated using leave-one-out cross-validation and 10-fold cross-validation. The results show that models using gait data from the instrumented footwear outperformed those based on traditional tests and walkway data, with area under the receiver operating characteristic curve (AUC) values for predicting prospective falls being 0.47, 0.66, and 0.80, respectively. The sensitivity of the models increases when they are trained using both past and future falls data, rather than relying solely on past or future falls data. This study demonstrates the potential of instrumented footwear for fall risk assessment in elderly individuals. The findings provide valuable insights for fall prevention and care, highlighting the superior predictive capabilities of the developed system compared to traditional methods.

Radar-Based Fall Detection: A Survey

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

Low-Cost Non-Wearable Fall Detection System Implemented on a Single Board Computer for People in Need of Care

This work aims at proposing an affordable, non-wearable system to detect falls of people in need of care. The proposal uses artificial vision based on deep learning techniques implemented on a Raspberry Pi4 4GB RAM with a High-Definition IR-CUT camera. The CNN architecture classifies detected people into five classes: fallen, crouching, sitting, standing, and lying down. When a fall is detected, the system sends an alert notification to mobile devices through the Telegram instant messaging platform. The system was evaluated considering real daily indoor activities under different conditions: outfit, lightning, and distance from camera. Results show a good trade-off between performance and cost of the system. Obtained performance metrics are: precision of 96.4%, specificity of 96.6%, accuracy of 94.8%, and sensitivity of 93.1%. Regarding privacy concerns, even though this system uses a camera, the video is not recorded or monitored by anyone, and pictures are only sent in case of fall detection. This work can contribute to reducing the fatal consequences of falls in people in need of care by providing them with prompt attention. Such a low-cost solution would be desirable, particularly in developing countries with limited or no medical alert systems and few resources.

Human Fall Detection with Ultra-Wideband Radar and Adaptive Weighted Fusion

To address the challenges in recognizing various types of falls, which often exhibit high similarity and are difficult to distinguish, this paper proposes a human fall classification system based on the SE-Residual Concatenate Network (SE-RCNet) with adaptive weighted fusion. First, we designed the innovative SE-RCNet network, incorporating SE modules after dense and residual connections to automatically recalibrate feature channel weights and suppress irrelevant features. Subsequently, this network was used to train and classify three types of radar images: time-distance images, time-distance images, and distance-distance images. By adaptively fusing the classification results of these three types of radar images, we achieved higher action recognition accuracy. Experimental results indicate that SE-RCNet achieved F1-scores of 94.0%, 94.3%, and 95.4% for the three radar image types on our self-built dataset. After applying the adaptive weighted fusion method, the F1-score further improved to 98.1%.

An Effective Deep Learning Framework for Fall Detection: Model Development and Study Design

BACKGROUND

Fall detection is of great significance in safeguarding human health. By monitoring the motion data, a fall detection system (FDS) can detect a fall accident. Recently, wearable sensors-based FDSs have become the mainstream of research, which can be categorized into threshold-based FDSs using experience, machine learning-based FDSs using manual feature extraction, and deep learning (DL)-based FDSs using automatic feature extraction. However, most FDSs focus on the global information of sensor data, neglecting the fact that different segments of the data contribute variably to fall detection. This shortcoming makes it challenging for FDSs to accurately distinguish between similar human motion patterns of actual falls and fall-like actions, leading to a decrease in detection accuracy.

OBJECTIVE

This study aims to develop and validate a DL framework to accurately detect falls using acceleration and gyroscope data from wearable sensors. We aim to explore the essential contributing features extracted from sensor data to distinguish falls from activities of daily life. The significance of this study lies in reforming the FDS by designing a weighted feature representation using DL methods to effectively differentiate between fall events and fall-like activities.

METHODS

Based on the 3-axis acceleration and gyroscope data, we proposed a new DL architecture, the dual-stream convolutional neural network self-attention (DSCS) model. Unlike previous studies, the used architecture can extract global feature information from acceleration and gyroscope data. Additionally, we incorporated a self-attention module to assign different weights to the original feature vector, enabling the model to learn the contribution effect of the sensor data and enhance classification accuracy. The proposed model was trained and tested on 2 public data sets: SisFall and MobiFall. In addition, 10 participants were recruited to carry out practical validation of the DSCS model. A total of 1700 trials were performed to test the generalization ability of the model.

RESULTS

The fall detection accuracy of the DSCS model was 99.32% (recall=99.15%; precision=98.58%) and 99.65% (recall=100%; precision=98.39%) on the test sets of SisFall and MobiFall, respectively. In the ablation experiment, we compared the DSCS model with state-of-the-art machine learning and DL models. On the SisFall data set, the DSCS model achieved the second-best accuracy; on the MobiFall data set, the DSCS model achieved the best accuracy, recall, and precision. In practical validation, the accuracy of the DSCS model was 96.41% (recall=95.12%; specificity=97.55%).

CONCLUSIONS

This study demonstrates that the DSCS model can significantly improve the accuracy of fall detection on 2 publicly available data sets and performs robustly in practical validation.

The effectiveness of a fall detection device in older nursing home residents: a pilot study

BACKGROUND

Real-world research to evaluate the effect of device technology in preventing fall-related morbidity is limited. This pilot study aims to investigate the effectiveness of a non-wearable fall detection device in older nursing home residents.

METHODS

The study was conducted in a nursing home with single-resident rooms. Fall detection devices were randomly set up in half of the rooms. Demographic data, comorbidities, lists of medications, and functional, nutritional, and frailty status were recorded. The residents were followed up for 3 months. The primary outcome was falls and the secondary outcome was all-cause mortality.

RESULTS

A total of 26 participants were enrolled in the study. The study group consisted of 13 residents who had a fall detection device in their rooms. The remaining 13 residents on the same floor formed the control group. Participants had a mean age of 82 ± 10 years and 89% of the residents were female. The most prevalent comorbidity was dementia. Two residents from the control group and one resident from the study group experienced a fall event during follow-up. The fall events in the control group were identified retrospectively by the nursing home staff, whereas the fall in the study group received a prompt response from the staff who were notified by the alarm. One resident was transferred to the hospital and died due to a non-fall related reason.

CONCLUSION

Device technology may provide an opportunity for timely intervention to prevent fall-related morbidity in institutionalized older adults.

Data Interoperability for Ambulatory Monitoring of Cardiovascular Disease: A Scientific Statement From the American Heart Association

Wearable devices are increasingly used by a growing portion of the population to track health and illnesses. The data emerging from these devices can potentially transform health care. This requires an interoperability framework that enables the deployment of platforms, sensors, devices, and software applications within diverse health systems, aiming to facilitate innovation in preventing and treating cardiovascular disease. However, the current data ecosystem includes several noninteroperable systems that inhibit such objectives. The design of clinically meaningful systems for accessing and incorporating these data into clinical workflows requires strategies to ensure the quality of data and clinical content and patient and caregiver accessibility. This scientific statement aims to address the best practices, gaps, and challenges pertaining to data interoperability in this area, with considerations for (1) data integration and the scope of measures, (2) application of these data into clinical approaches/strategies, and (3) regulatory/ethical/legal issues.

Integrating attention mechanism and multi-scale feature extraction for fall detection

Addressing the critical need for accurate fall event detection due to their potentially severe impacts, this paper introduces the Spatial Channel and Pooling Enhanced You Only Look Once version 5 small (SCPE-YOLOv5s) model. Fall events pose a challenge for detection due to their varying scales and subtle pose features. To address this problem, SCPE-YOLOv5s introduces spatial attention to the Efficient Channel Attention (ECA) network, which significantly enhances the model's ability to extract features from spatial pose distribution. Moreover, the model integrates average pooling layers into the Spatial Pyramid Pooling (SPP) network to support the multi-scale extraction of fall poses. Meanwhile, by incorporating the ECA network into SPP, the model effectively combines global and local features to further enhance the feature extraction. This paper validates the SCPE-YOLOv5s on a public dataset, demonstrating that it achieves a mean Average Precision of 88.29 %, outperforming the You Only Look Once version 5 small by 4.87 %. Additionally, the model achieves 57.4 frames per second. Therefore, SCPE-YOLOv5s provides a novel solution for fall event detection.

Edge Computing Transformers for Fall Detection in Older Adults

The global trend of increasing life expectancy introduces new challenges with far-reaching implications. Among these, the risk of falls among older adults is particularly significant, affecting individual health and the quality of life, and placing an additional burden on healthcare systems. Existing fall detection systems often have limitations, including delays due to continuous server communication, high false-positive rates, low adoption rates due to wearability and comfort issues, and high costs. In response to these challenges, this work presents a reliable, wearable, and cost-effective fall detection system. The proposed system consists of a fit-for-purpose device, with an embedded algorithm and an Inertial Measurement Unit (IMU), enabling real-time fall detection. The algorithm combines a Threshold-Based Algorithm (TBA) and a neural network with low number of parameters based on a Transformer architecture. This system demonstrates notable performance with 95.29% accuracy, 93.68% specificity, and 96.66% sensitivity, while only using a 0.38% of the trainable parameters used by the other approach.

Multimodal fall detection for solitary individuals based on audio-video decision fusion processing

Falls often pose significant safety risks to solitary individuals, especially the elderly. Implementing a fast and efficient fall detection system is an effective strategy to address this hidden danger. We propose a multimodal method based on audio and video. On the basis of using non-intrusive equipment, it reduces to a certain extent the false negative situation that the most commonly used video-based methods may face due to insufficient lighting conditions, exceeding the monitoring range, etc. Therefore, in the foreseeable future, methods based on audio and video fusion are expected to become the best solution for fall detection. Specifically, this article outlines the following methodology: the video-based model utilizes YOLOv7-Pose to extract key skeleton joints, which are then fed into a two stream Spatial Temporal Graph Convolutional Network (ST-GCN) for classification. Meanwhile, the audio-based model employs log-scaled mel spectrograms to capture different features, which are processed through the MobileNetV2 architecture for detection. The final decision fusion of the two results is achieved through linear weighting and Dempster-Shafer (D-S) theory. After evaluation, our multimodal fall detection method significantly outperforms the single modality method, especially the evaluation metric sensitivity increased from 81.67% in single video modality to 96.67% (linear weighting) and 97.50% (D-S theory), which emphasizing the effectiveness of integrating video and audio data to achieve more powerful and reliable fall detection in complex and diverse daily life environments.

Development of artificial intelligence edge computing based wearable device for fall detection and prevention of elderly people

Elderly falls are a major concerning threat resulting in over 1.5-2 million elderly people experiencing severe injuries and 1 million deaths yearly. Falls experienced by Elderly people may lead to a long-term negative impact on their physical and psychological health conditions. Major healthcare research had focused on this lately to detect and prevent the fall. In this work, an Artificial Intelligence (AI) edge computing based wearable device is designed and developed for detection and prevention of fall of elderly people. Further, the various deep learning algorithms such as Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU) are utilized for activity recognition of elderly. Also, the CNN-LSTM, RNN-LSTM and GRU-LSTM with and without attention layer respectively are utilized and the performance metrics are analyzed to find the best deep learning model. Furthermore, the three different hardware boards such as Jetson Nano developer board, Raspberry PI 3 and 4 are utilized as an AI edge computing device and the best deep learning model is implemented and the computation time is evaluated. Results demonstrate that the CNN-LSTM with attention layer exhibits the accuracy, recall, precision and F1_Score of 97%, 98%, 98% and 0.98 respectively which is better when compared to other deep learning models. Also, the computation time of NVIDIA Jetson Nano is less when compared to other edge computing devices. This work appears to be of high societal relevance since the proposed wearable device can be used to monitor the activity of elderly and prevents the elderly falls which improve the quality of life of elderly people.

Synthetic IMU Datasets and Protocols Can Simplify Fall Detection Experiments and Optimize Sensor Configuration

Falls represent a significant cause of injury among the elderly population. Extensive research has been devoted to the utilization of wearable IMU sensors in conjunction with machine learning techniques for fall detection. To address the challenge of acquiring costly training data, this paper presents a novel method that generates a substantial volume of synthetic IMU data with minimal actual fall experiments. First, unmarked 3D motion capture technology is employed to reconstruct human movements. Subsequently, utilizing the biomechanical simulation platform Opensim and forward kinematic methods, an ample amount of training data from various body segments can be custom generated. Synthetic IMU data was then used to train a machine learning model, achieving testing accuracies of 91.99% and 86.62% on two distinct datasets of actual fall-related IMU data. Building upon the simulation framework, this paper further optimized the single IMU attachment position and multiple IMU combinations on fall detection. The proposed method simplifies fall detection data acquisition experiments, provides novel venue for generating low cost synthetic data in scenario where acquiring data for machine learning is challenging and paves the way for customizing machine learning configurations.

Reliable relay assisted communications for IoT based fall detection

Robust wireless communication using relaying system and Non-Orthogonal Multiple Access (NOMA) will be extensively used for future IoT applications. In this paper, we consider a fall detection IoT application in which elderly patients are equipped with wearable motion sensors. Patient motion data is sent to fog data servers via a NOMA-based relaying system, thereby improving the communication reliability. We analyze the average signal-to-interference-plus-noise (SINR) performance of the NOMA-based relaying system, where the source node transmits two different symbols to the relay and destination node by employing superposition coding over Rayleigh fading channels. In the amplify-and-forward (AF) based relaying, the relay re-transmits the received signal after amplification, whereas, in the decode-and-forward (DF) based relaying, the relay only re-transmits the symbol having lower NOMA power coefficient. We derive closed-form average SINR expressions for AF and DF relaying systems using NOMA. The average SINR expressions for AF and DF relaying systems are derived in terms of computationally efficient functions, namely Tricomi confluent hypergeometric and Meijer's G functions. Through simulations, it is shown that the average SINR values computed using the derived analytical expressions are in excellent agreement with the simulation-based average SINR results.

Enhancing elderly care: Efficient and reliable real-time fall detection algorithm

Background and Objective

Falls pose a significant risk to public health, especially for the elderly population, and could potentially result in severe injuries or even death. A reliable fall detection system is urgently needed to recognise and promptly alert to falls effectively. A vision-based fall detection system has the advantage of being non-invasive and affordable compared with another popular approach using wearable sensors. Nevertheless, the present challenge lies in the algorithm's limited on-device operating speed due to extremely high computational demands, and the high computational demands are usually essential to improve the performance for the complex scene. Therefore, it is crucial to address the above challenge in computational power and complex scenes.

Methods

This article presents the implementation of a real-time fall detection algorithm with low computational costs using a single webcam. The suggested method optimises precision and efficiency by synthesising the strengths of background subtraction and the human pose estimation model BlazePose. The biomechanical features, derived from body key points identified by BlazePose, are utilised in a random forest model for classifying fall events.

Results

The proposed algorithm achieves 89.99% accuracy and 29.7 FPS with a laptop CPU on the UR Fall Detection dataset and the Le2i Fall Detection dataset. The algorithm shows great generalisation and robustness in different scenarios.

Conclusion

Due to the low computational power of the system, the findings also suggest the potential for implementing the system in small-scale medical monitoring equipment, which maximises its practical value in digital health.

Survey of Transfer Learning Approaches in the Machine Learning of Digital Health Sensing Data

Machine learning and digital health sensing data have led to numerous research achievements aimed at improving digital health technology. However, using machine learning in digital health poses challenges related to data availability, such as incomplete, unstructured, and fragmented data, as well as issues related to data privacy, security, and data format standardization. Furthermore, there is a risk of bias and discrimination in machine learning models. Thus, developing an accurate prediction model from scratch can be an expensive and complicated task that often requires extensive experiments and complex computations. Transfer learning methods have emerged as a feasible solution to address these issues by transferring knowledge from a previously trained task to develop high-performance prediction models for a new task. This survey paper provides a comprehensive study of the effectiveness of transfer learning for digital health applications to enhance the accuracy and efficiency of diagnoses and prognoses, as well as to improve healthcare services. The first part of this survey paper presents and discusses the most common digital health sensing technologies as valuable data resources for machine learning applications, including transfer learning. The second part discusses the meaning of transfer learning, clarifying the categories and types of knowledge transfer. It also explains transfer learning methods and strategies, and their role in addressing the challenges in developing accurate machine learning models, specifically on digital health sensing data. These methods include feature extraction, fine-tuning, domain adaptation, multitask learning, federated learning, and few-/single-/zero-shot learning. This survey paper highlights the key features of each transfer learning method and strategy, and discusses the limitations and challenges of using transfer learning for digital health applications. Overall, this paper is a comprehensive survey of transfer learning methods on digital health sensing data which aims to inspire researchers to gain knowledge of transfer learning approaches and their applications in digital health, enhance the current transfer learning approaches in digital health, develop new transfer learning strategies to overcome the current limitations, and apply them to a variety of digital health technologies.

Recurrent Neural Network Methods for Extracting Dynamic Balance Variables during Gait from a Single Inertial Measurement Unit

Monitoring dynamic balance during gait is critical for fall prevention in the elderly. The current study aimed to develop recurrent neural network models for extracting balance variables from a single inertial measurement unit (IMU) placed on the sacrum during walking. Thirteen healthy young and thirteen healthy older adults wore the IMU during walking and the ground truth of the inclination angles (IA) of the center of pressure to the center of mass vector and their rates of changes (RCIA) were measured simultaneously. The IA, RCIA, and IMU data were used to train four models (uni-LSTM, bi-LSTM, uni-GRU, and bi-GRU), with 10% of the data reserved to evaluate the model errors in terms of the root-mean-squared errors (RMSEs) and percentage relative RMSEs (rRMSEs). Independent t-tests were used for between-group comparisons. The sensitivity, specificity, and Pearson's r for the effect sizes between the model-predicted data and experimental ground truth were also obtained. The bi-GRU with the weighted MSE model was found to have the highest prediction accuracy, computational efficiency, and the best ability in identifying statistical between-group differences when compared with the ground truth, which would be the best choice for the prolonged real-life monitoring of gait balance for fall risk management in the elderly.

AI-Enabled Public Surveillance Cameras for Rapid Emergency Medical Service Activation in Out-of-Hospital Cardiac Arrests

This study aims to evaluate the potential usefulness of a novel artificial intelligence (AI)-based video processing algorithm for rapidly activating ambulance services (EMS) in unwitnessed out-of-hospital cardiac arrest (OHCA) cases in public places. We hypothesized that AI should activate EMS using public surveillance cameras after detecting a person fallen due to OHCA. We created an AI model based on our experiment performed at the Lithuanian University of Health Sciences, Kaunas, Lithuania, in Spring 2023. Our research highlights the potential benefits of AI-based surveillance cameras for rapidly detecting and activating EMS during cardiac arrests.

Ambient assisted living for frail people through human activity recognition: state-of-the-art, challenges and future directions

Ambient Assisted Living is a concept that focuses on using technology to support and enhance the quality of life and well-being of frail or elderly individuals in both indoor and outdoor environments. It aims at empowering individuals to maintain their independence and autonomy while ensuring their safety and providing assistance when needed. Human Activity Recognition is widely regarded as the most popular methodology within the field of Ambient Assisted Living. Human Activity Recognition involves automatically detecting and classifying the activities performed by individuals using sensor-based systems. Researchers have employed various methodologies, utilizing wearable and/or non-wearable sensors, and employing algorithms ranging from simple threshold-based techniques to more advanced deep learning approaches. In this review, literature from the past decade is critically examined, specifically exploring the technological aspects of Human Activity Recognition in Ambient Assisted Living. An exhaustive analysis of the methodologies adopted, highlighting their strengths and weaknesses is provided. Finally, challenges encountered in the field of Human Activity Recognition for Ambient Assisted Living are thoroughly discussed. These challenges encompass issues related to data collection, model training, real-time performance, generalizability, and user acceptance. Miniaturization, unobtrusiveness, energy harvesting and communication efficiency will be the crucial factors for new wearable solutions.

PIPTO: Precise Inertial-Based Pipeline for Threshold-Based Fall Detection Using Three-Axis Accelerometers

After traffic-related incidents, falls are the second cause of human death, presenting the highest percentage among the elderly. Aiming to address this problem, the research community has developed methods built upon different sensors, such as wearable, ambiance, or hybrid, and various techniques, such as those that are machine learning- and heuristic based. Concerning the models used in the former case, they classify the input data between fall and no fall, and specific data dimensions are required. Yet, when algorithms that adopt heuristic techniques, mainly using thresholds, are combined with the previous models, they reduce the computational cost. To this end, this article presents a pipeline for detecting falls through a threshold-based technique over the data provided by a three-axis accelerometer. This way, we propose a low-complexity system that can be adopted from any acceleration sensor that receives information at different frequencies. Moreover, the input lengths can differ, while we achieve to detect multiple falls in a time series of sum vector magnitudes, providing the specific time range of the fall. As evaluated on several datasets, our pipeline reaches high performance results at 90.40% and 91.56% sensitivity on MMsys and KFall, respectively, while the generated specificity is 93.96% and 85.90%. Lastly, aiming to facilitate the research community, our framework, entitled PIPTO (drawing inspiration from the Greek verb "πι´πτω", signifying "to fall"), is open sourced in Python and C.

Comparison of machine learning approaches for near-fall-detection with motion sensors

Introduction

Falls are one of the most common causes of emergency hospital visits in older people. Early recognition of an increased fall risk, which can be indicated by the occurrence of near-falls, is important to initiate interventions.

Methods

In a study with 87 subjects we simulated near-fall events on a perturbation treadmill and recorded them with inertial measurement units (IMU) at seven different positions. We investigated different machine learning models for the near-fall detection including support vector machines, AdaBoost, convolutional neural networks, and bidirectional long short-term memory networks. Additionally, we analyzed the influence of the sensor position on the classification results.

Results

The best results showed a DeepConvLSTM with an F1 score of 0.954 (precision 0.969, recall 0.942) at the sensor position "left wrist."

Discussion

Since these results were obtained in the laboratory, the next step is to evaluate the suitability of the classifiers in the field.

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.

An Indoor Fall Monitoring System: Robust, Multistatic Radar Sensing and Explainable, Feature-Resonated Deep Neural Network

Indoor fall monitoring is challenging for community-dwelling older adults due to the need for high accuracy and privacy concerns. Doppler radar is promising, given its low-cost and contactless sensing mechanism. However, the line-of-sight restriction limits the application of radar sensing in practice, as the Doppler signature will vary when the sensing angle changes, and signal strength will substantially degrade with large aspect angles. Additionally, the similarity of the Doppler signatures among different fall types makes classification challenging. To address these problems, we first present an experimental study to obtain Doppler signals under large and arbitrary aspect angles for diverse types of simulated activities. We then develop a novel, explainable, multi-stream, feature-resonated neural network (eMSFRNet) that achieves fall detection and a pioneering study of classifying seven fall types. eMSFRNet is robust to radar sensing angles and subjects, and is the first method that can resonate and enhance feature information from noisy/weak Doppler signatures. The multiple feature extractors - from ResNet, DenseNet, and VGGNet - extract diverse feature information with various spatial abstractions from a pair of Doppler signals. The resonated-fusion translates the multi-stream features to a single salient feature that is critical to fall detection and classification. eMSFRNet achieved 99.3% accuracy detecting falls and 76.8% accuracy classifying seven fall types. Our work is the firstmultistatic robust sensing system that overcomes the challenges associated with Doppler signatures under large and arbitrary aspect angles. Our work also demonstrates the potential to accommodate radar monitoring tasks that demand precise and robust sensing.

Energy-Aware IoT-Based Method for a Hybrid On-Wrist Fall Detection System Using a Supervised Dictionary Learning Technique

In recent decades, falls have posed multiple critical health issues, especially for the older population, with their emerging growth. Recent research has shown that a wrist-based fall detection system offers an accessory-like comfortable solution for Internet of Things (IoT)-based monitoring. Nevertheless, an autonomous device for anywhere-anytime may present an energy consumption concern. Hence, this paper proposes a novel energy-aware IoT-based architecture for Message Queuing Telemetry Transport (MQTT)-based gateway-less monitoring for wearable fall detection. Accordingly, a hybrid double prediction technique based on Supervised Dictionary Learning was implemented to reinforce the detection efficiency of our previous works. A controlled dataset was collected for training (offline), while a real set of measurements of the proposed system was used for validation (online). It achieved a noteworthy offline and online detection performance of 99.8% and 91%, respectively, overpassing most of the related works using only an accelerometer. In the worst case, the system showed a battery consumption optimization by a minimum of 27.32 working hours, significantly higher than other research prototypes. The approach presented here proves to be promising for real applications, which require a reliable and long-term anywhere-anytime solution.

SHMAD: A Smart Health Care System to Monitor Alzheimer's Disease Patients

BACKGROUND

In the digital era monitoring the patient's health status is more effective and consistent with smart healthcare systems. Smart health care facilitates secure and reliable maintenance of patient data. Sensors, machine learning algorithms, Internet of things, and wireless technology has led to the development of Artificial Intelligence-driven Internet of Things models.

OBJECTIVE

This research study proposes an Artificial Intelligence driven Internet of Things model to monitor Alzheimer's disease patient condition. The proposed Smart health care system to monitor and alert caregivers of Alzheimer's disease patients includes different modules to monitor the health parameters of the patients. This study implements the detection of fall episodes using an artificial intelligence model in Python.

METHODS

The fall detection model is implemented with data acquired from the IMU open dataset. The ensemble machine learning algorithm AdaBoost performs classification of the fall episode and daily life activity using the feature set of each data sample. The common machine learning classification algorithms are compared for their performance on the IMU fall dataset.

RESULTS

AdaBoost ensemble classifier exhibits high performance compared to the other machine learning algorithms. The AdaBoost classifier shows 100% accuracy for the IMU dataset. This high accuracy is achieved as multiple weak learners in the ensemble model classify the data samples in the test data accurately.

CONCLUSIONS

This study proposes a smart healthcare system for monitoring Alzheimer's disease patients. The proposed model can alert the caregiver in case of fall detection via mobile applications installed in smart devices.

Artificial intelligence for falls management in older adult care: A scoping review of nurses' role

AIM

This study aims to synthesize evidence on nurses' involvement in artificial intelligence research for managing falls in older adults.

BACKGROUND

Artificial intelligence techniques are used to analyse health datasets to aid clinical decision making, patient care and service delivery but nurses' involvement in this area of research for managing falls in older adults remains unknown.

EVALUATION

A scoping review was conducted. CINAHL, the Cochrane Library, Embase, MEDLI and PubMed were searched. Results were screened against inclusion criteria. Relevant data were extracted, and studies summarized using a descriptive approach.

KEY ISSUES

The evidence shows many artificial intelligence techniques, particularly machine learning, are used to identify falls risk factors and build predictive models that could help prevent falls in older adults, with nurses leading and participating in this research.

CONCLUSION

Further rigorous experimental research is needed to determine the effectiveness of algorithms in predicting aspects of falls in older adults and how to implement artificial intelligence tools in gerontological nursing practice.

IMPLICATIONS FOR NURSING MANAGEMENT

Nurses should pursue interdisciplinary collaborations and educational opportunities in artificial intelligence, so they can actively contribute to research on falls management. Nurses should facilitate the collection of digital falls datasets to support this emerging research agenda and the care of older adults.

Artificial intelligence assisted acute patient journey

Artificial intelligence is taking the world by storm and soon will be aiding patients in their journey at the hospital. The trials and tribulations of the healthcare system during the COVID-19 pandemic have set the stage for shifting healthcare from a physical to a cyber-physical space. A physician can now remotely monitor a patient, admitting them only if they meet certain thresholds, thereby reducing the total number of admissions at the hospital. Coordination, communication, and resource management have been core issues for any industry. However, it is most accurate in healthcare. Both systems and providers are exhausted under the burden of increasing data and complexity of care delivery, increasing costs, and financial burden. Simultaneously, there is a digital transformation of healthcare in the making. This transformation provides an opportunity to create systems of care that are artificial intelligence-enabled. Healthcare resources can be utilized more justly. The wastage of financial and intellectual resources in an overcrowded healthcare system can be avoided by implementing IoT, telehealth, and AI/ML-based algorithms. It is imperative to consider the design principles of the patient's journey while simultaneously prioritizing a better user experience to alleviate physician concerns. This paper discusses the entire blueprint of the AI/ML-assisted patient journey and its impact on healthcare provision.

A Simplistic and Cost-Effective Design for Real-World Development of an Ambient Assisted Living System for Fall Detection and Indoor Localization: Proof-of-Concept

Falls, highly common in the constantly increasing global aging population, can have a variety of negative effects on their health, well-being, and quality of life, including restricting their capabilities to conduct activities of daily living (ADLs), which are crucial for one’s sustenance. Timely assistance during falls is highly necessary, which involves tracking the indoor location of the elderly during their diverse navigational patterns associated with different activities to detect the precise location of a fall. With the decreasing caregiver population on a global scale, it is important that the future of intelligent living environments can detect falls during ADLs while being able to track the indoor location of the elderly in the real world. Prior works in these fields have several limitations, such as the lack of functionalities to detect falls and indoor locations in a simultaneous manner, high cost of implementation, complicated design, the requirement of multiple hardware components for deployment, and the necessity to develop new hardware for implementation, which make the wide-scale deployment of such technologies challenging. To address these challenges, this work proposes a cost-effective and simplistic design paradigm for an ambient assisted living system that can capture multimodal components of user behaviors during ADLs that are necessary for performing fall detection and indoor localization in a simultaneous manner in the real-world. Proof-of-concept results from real-world experiments are presented to uphold the effective working of the system. The findings from two comparative studies with prior works in this field are also presented to uphold the novelty of this work. The first comparative study shows how the proposed system outperforms prior works in the areas of indoor localization and fall detection in terms of the effectiveness of its software design and hardware design. The second comparative study shows that the cost of the development of this system is the lowest as compared to prior works in these fields, which involved real-world development of the underlining systems, thereby upholding its cost-effective nature.

Using Deep Neural Networks for Human Fall Detection Based on Pose Estimation

Requests for caring for and monitoring the health and safety of older adults are increasing nowadays and form a topic of great social interest. One of the issues that lead to serious concerns is human falls, especially among aged people. Computer vision techniques can be used to identify fall events, and Deep Learning methods can detect them with optimum accuracy. Such imaging-based solutions are a good alternative to body-worn solutions. This article proposes a novel human fall detection solution based on the Fast Pose Estimation method. The solution uses Time-Distributed Convolutional Long Short-Term Memory (TD-CNN-LSTM) and 1Dimentional Convolutional Neural Network (1D-CNN) models, to classify the data extracted from image frames, and achieved high accuracies: 98 and 97% for the 1D-CNN and TD-CNN-LSTM models, respectively. Therefore, by applying the Fast Pose Estimation method, which has not been used before for this purpose, the proposed solution is an effective contribution to accurate human fall detection, which can be deployed in edge devices due to its low computational and memory demands.

Relationship between Associated Neuropsychological Factors and Fall Risk Factors in Community-Dwelling Elderly

This study examined whether neuropsychological factors could affect fall risk factors in the community-dwelling elderly via correlation analysis. A total of 393 older adults (76.69 ± 6.01) participated in this study. Cognitive function, depression, fall efficacy, balance confidence, balance, gait, and muscle strength were evaluated, and the correlation between psychological factors and fall risk factors was analyzed. Additionally, a multiple regression analysis was conducted to determine whether or not there was a significant effect between psychological factors and fall risk factors. Analysis showed that the psychological factors examined were all significantly correlated with the fall risk factors. A correlation analysis between cognitive function and fall risk factors showed that the correlation coefficient of the 6-Meter Walk Test was highest; for depression and fall risk factors, the correlation coefficient of gait speed was highest; for fall efficacy and fall risk factors, the correlation coefficient of the 6-Meter Walk Test was highest; and for confidence in balancing and fall risk factors, the correlation coefficient of the 6-Meter Walk Test was highest. This study suggests that psychological factors affect fall risk factors in the community-dwelling elderly, and a multifaceted approach that includes psychological factors would be helpful in providing interventions for falls.

Denetimsiz Anomali Tespiti Yaklaşımı ile Düşme Algılama

Yaşlı nüfusunun hızla artması ve yaşlılığa bağlı olarak karşılaşılan fiziksel, duyusal ve bilişsel gerilemeler, düşmeyi her geçen gün büyüyen bir problem olarak karşımıza çıkarmakta ve düşme tespiti çalışmalarının hız kazanmasına sebep olmaktadır. Günlük aktivitelerin düşmeden ayırt edilmesinden ibaret olan düşme tespiti probleminde, denetimli öğrenme yaklaşımları kullanılmasına rağmen, düşmenin nadir rastlanan ve çok farklı biçimlerde karşılaşılabilen bir olay olması genel bir model elde edilmesine izin vermemektedir. Bu çalışmada denetimsiz anomali tespiti ile düşmenin belirlenmesi önerilmektedir. Denetimsiz öğrenme modelinin elde edilmesinde ve model vasıtasıyla düşmenin tespitinde 35 tip düşme ve 44 tip günlük aktiviteye sahip kapsamlı bir veri setinden faydalanılmıştır. Denetimsiz öğrenme yöntemi olan Gauss karışım modelinin eğitiminde, günlük aktivitelerden toplanan 3-eksen ivmeölçer sinyallerinden elde edilen öznitelikler kullanılmıştır. Test aşamasında model, düşme ve günlük aktivite verileri ile karşılaşmış, modele göre olasılığı çok düşük olan veriler anomali, dolayısıyla düşme olarak kabul edilmiştir. Testlerde düşmeler %90,5 civarında doğru olarak tespit edilmiştir. Sonuçlar düşmenin anomali tespiti yaklaşımları ile belirlenebileceğini ve makine öğrenmesi modelinin elde edilmesi için yalnız günlük aktivite verilerinin yeterli olduğu yaklaşımını doğrulamaktadır.

Exploring the impact of experiencing a long lie fall on physical and clinical outcomes in older people requiring an ambulance: A systematic review

BACKGROUND

The long term impacts of experiencing a 'long lie' following a fall in the older person are poorly understood. This systematic review explored the impact of a long lie fall on physical and clinical outcomes in older people requiring an ambulance.

METHODS

PRISMA guidelines were followed.

RESULTS

70 studies were identified. Nine studies were suitable for full review. Four studies meeting the inclusion criteria were included. The Critical Appraisal Skill Programme (CASP) assessed the quality of all included studies. Three studies reported on people aged 65 years and older. One study reported on people aged over 90 years. Personal alarm use was examined in two studies. One study explored patient characteristics of people confirmed to have fallen by paramedics at the scene. One study examined re-contact and characteristics of fallers referred to a falls prevention service.

DISCUSSION

Cognitive impairment and long lie were a caveat for falls and repeated falls. Personal alarm use was infrequent, suggesting a need for supporting the older patient in appropriate alarm use and exploration of newer technologies to alleviate their need. Future research should focus on interventions for wearable, smart and e-technology for automatic fall detection and qualitative exploration of the lived experience.

Machine Learning-Based Classification of Human Behaviors and Falls in Restroom via Dual Doppler Radar Measurements

This study presents a radar-based remote measurement system for classification of human behaviors and falls in restrooms without privacy invasion. Our system uses a dual Doppler radar mounted onto a restroom ceiling and wall. Machine learning methods, including the convolutional neural network (CNN), long short-term memory, support vector machine, and random forest methods, are applied to the Doppler radar data to verify the model’s efficiency and features. Experimental results from 21 participants demonstrated the accurate classification of eight realistic behaviors, including falling. Using the Doppler spectrograms (time–velocity distribution) as the inputs, CNN showed the best results with an overall classification accuracy of 95.6% and 100% fall classification accuracy. We confirmed that these accuracies were better than those achieved by conventional restroom monitoring techniques using thermal sensors and radars. Furthermore, the comparison results of various machine learning methods and cases using each radar’s data show that the higher-order derivative parameters of acceleration and jerk, and the motion information in the horizontal direction are the efficient features for behavior classification in a restroom. These findings indicate that daily restroom monitoring using the proposed radar system accurately recognizes human behaviors and allows early detection of fall accidents.

An Instrumented Apartment to Monitor Human Behavior: A Pilot Case Study in the NeuroTec Loft

For patients suffering from neurodegenerative disorders, the behavior and activities of daily living are an indicator of a change in health status, and home-monitoring over a prolonged period of time by unobtrusive sensors is a promising technology to foster independent living and maintain quality of life. The aim of this pilot case study was the development of a multi-sensor system in an apartment to unobtrusively monitor patients at home during the day and night. The developed system is based on unobtrusive sensors using basic technologies and gold-standard medical devices measuring physiological (e.g., mobile electrocardiogram), movement (e.g., motion tracking system), and environmental parameters (e.g., temperature). The system was evaluated during one session by a healthy 32-year-old male, and results showed that the sensor system measured accurately during the participant’s stay. Furthermore, the participant did not report any negative experiences. Overall, the multi-sensor system has great potential to bridge the gap between laboratories and older adults’ homes and thus for a deep and novel understanding of human behavioral and neurological disorders. Finally, this new understanding could be utilized to develop new algorithms and sensor systems to address problems and increase the quality of life of our aging society and patients with neurological disorders.

Cooperative ankle-exoskeleton control can reduce effort to recover balance after unexpected disturbances during walking

BACKGROUND

In the last two decades, lower-limb exoskeletons have been developed to assist human standing and locomotion. One of the ongoing challenges is the cooperation between the exoskeleton balance support and the wearer control. Here we present a cooperative ankle-exoskeleton control strategy to assist in balance recovery after unexpected disturbances during walking, which is inspired on human balance responses.

METHODS

We evaluated the novel controller in ten able-bodied participants wearing the ankle modules of the Symbitron exoskeleton. During walking, participants received unexpected forward pushes with different timing and magnitude at the pelvis level, while being supported (Exo-Assistance) or not (Exo-NoAssistance) by the robotic assistance provided by the controller. The effectiveness of the assistive strategy was assessed in terms of (1) controller performance (Detection Delay, Joint Angles, and Exerted Ankle Torques), (2) analysis of effort (integral of normalized Muscle Activity after perturbation onset); and (3) Analysis of center of mass COM kinematics (relative maximum COM Motion, Recovery Time and Margin of Stability) and spatio-temporal parameters (Step Length and Swing Time).

RESULTS

In general, the results show that when the controller was active, it was able to reduce participants' effort while keeping similar ability to counteract and withstand the balance disturbances. Significant reductions were found for soleus and gastrocnemius medialis activity of the stance leg when comparing Exo-Assistance and Exo-NoAssistance walking conditions.

CONCLUSIONS

The proposed controller was able to cooperate with the able-bodied participants in counteracting perturbations, contributing to the state-of-the-art of bio-inspired cooperative ankle exoskeleton controllers for supporting dynamic balance. In the future, this control strategy may be used in exoskeletons to support and improve balance control in users with motor disabilities.

A convolution neural network approach for fall detection based on adaptive channel selection of UWB radar signals

According to the World Health Organization and other authorities, falls are one of the main causes of accidental injuries among the elderly population. Therefore, it is essential to detect and predict the fall activities of older persons in indoor environments such as homes, nursing, senior residential centers, and care facilities. Due to non-contact and signal confidentiality characteristics, radar equipment is widely used in indoor care, detection, and rescue. This paper proposes an adaptive channel selection algorithm to separate the activity signals from the background using an ultra-wideband radar and to generalize fused features of frequency- and time-domain images which will be sent to a lightweight convolutional neural network to detect and recognize fall activities. The experimental results show that the method is able to distinguish three types of fall activities (i.e., stand to fall, bow to fall, and squat to fall) and obtain a high recognition accuracy up to 95.7%.

High-Efficiency Multi-Sensor System for Chair Usage Detection

Recognizing Activities of Daily Living (ADL) or detecting falls in domestic environments require monitoring the movements and positions of a person. Several approaches use wearable devices or cameras, especially for fall detection, but they are considered intrusive by many users. To support such activities in an unobtrusive way, ambient-based solutions are available (e.g., based on PIRs, contact sensors, etc.). In this paper, we focus on the problem of sitting detection exploiting only unobtrusive sensors. In fact, sitting detection can be useful to understand the position of the user in many activities of the daily routines. While identifying sitting/lying on a sofa or bed is reasonably simple with pressure sensors, detecting whether a person is sitting on a chair is an open problem due to the natural chair position volatility. This paper proposes a reliable, not invasive and energetically sustainable system that can be used on chairs already present in the home. In particular, the proposed solution fuses the data of an accelerometer and a capacitive coupling sensor to understand if a person is sitting or not, discriminating the case of objects left on the chair. The results obtained in a real environment setting show an accuracy of 98.6% and a precision of 95%.

Design and Implementation of a Wearable Accelerometer-Based Motion/Tilt Sensing Internet of Things Module and Its Application to Bed Fall Prevention

Accelerometer-based motion sensing has been extensively applied to fall detection. However, such applications can only detect fall accidents; therefore, a system that can prevent fall accidents is desirable. Bed falls account for more than half of patient falls and are preceded by a clear warning indicator: the patient attempting to get out of bed. This study designed and implemented an Internet of Things module, namely, Bluetooth low-energy-enabled Accelerometer-based Sensing In a Chip-packaging (BASIC) module, with a tilt-sensing algorithm based on the patented low-complexity COordinate Rotation DIgital Computer (CORDIC)-based algorithm for tilt angle conversions. It is applied for detecting the postural changes (from lying down to sitting up) and to protect individuals at a high risk of bed falls by prompting caregivers to take preventive actions and assist individuals trying to get up. This module demonstrates how motion and tilt sensing can be applied to bed fall prevention. The module can be further miniaturized or integrated into a wearable device and commercialized in smart health-care applications for bed fall prevention in hospitals and homes.

A Study of One-Class Classification Algorithms for Wearable Fall Sensors

In recent years, the popularity of wearable devices has fostered the investigation of automatic fall detection systems based on the analysis of the signals captured by transportable inertial sensors. Due to the complexity and variety of human movements, the detection algorithms that offer the best performance when discriminating falls from conventional Activities of Daily Living (ADLs) are those built on machine learning and deep learning mechanisms. In this regard, supervised machine learning binary classification methods have been massively employed by the related literature. However, the learning phase of these algorithms requires mobility patterns caused by falls, which are very difficult to obtain in realistic application scenarios. An interesting alternative is offered by One-Class Classifiers (OCCs), which can be exclusively trained and configured with movement traces of a single type (ADLs). In this paper, a systematic study of the performance of various typical OCCs (for diverse sets of input features and hyperparameters) is performed when applied to nine public repositories of falls and ADLs. The results show the potentials of these classifiers, which are capable of achieving performance metrics very similar to those of supervised algorithms (with values for the specificity and the sensitivity higher than 95%). However, the study warns of the need to have a wide variety of types of ADLs when training OCCs, since activities with a high degree of mobility can significantly increase the frequency of false alarms (ADLs identified as falls) if not considered in the data subsets used for training.

A Study of Fall Detection in Assisted Living: Identifying and Improving the Optimal Machine Learning Method

This paper makes four scientific contributions to the field of fall detection in the elderly to contribute to their assisted living in the future of Internet of Things (IoT)-based pervasive living environments, such as smart homes. First, it presents and discusses a comprehensive comparative study, where 19 different machine learning methods were used to develop fall detection systems, to deduce the optimal machine learning method for the development of such systems. This study was conducted on two different datasets, and the results show that out of all the machine learning methods, the k-NN classifier is best suited for the development of fall detection systems in terms of performance accuracy. Second, it presents a framework that overcomes the limitations of binary classifier-based fall detection systems by being able to detect falls and fall-like motions. Third, to increase the trust and reliance on fall detection systems, it introduces a novel methodology based on the usage of k-folds cross-validation and the AdaBoost algorithm that improves the performance accuracy of the k-NN classifier-based fall detection system to the extent that it outperforms all similar works in this field. This approach achieved performance accuracies of 99.87% and 99.66%, respectively, when evaluated on the two datasets. Finally, the proposed approach is also highly accurate in detecting the activity of standing up from a lying position to infer whether a fall was followed by a long lie, which can cause minor to major health-related concerns. The above contributions address multiple research challenges in the field of fall detection, that we identified after conducting a comprehensive review of related works, which is also presented in this paper.

High Accuracy WiFi-Based Human Activity Classification System with Time-Frequency Diagram CNN Method for Different Places

Older people are very likely to fall, which is a significant threat to the health. However, falls are preventable and are not necessarily an inevitable part of aging. Many different fall detection systems have been developed to help people avoid falling. However, traditional systems based on wearable devices or image recognition-based have many disadvantages, such as user-unfriendly, privacy issues. Recently, WiFi-based fall detection systems try to solve the above problems. However, there is a common problem of reduced accuracy. Since the system is trained at the original signal collecting/training place, however, the application is at a different place. The proposed solution only extracts the features of the changed signal, which is caused by a specific human action. To implement this, we used Channel State Information (CSI) to train Convolutional Neural Networks (CNNs) and further classify the action. We have designed a prototype to test the performance of our proposed method. Our simulation results show an average accuracy of same place and different place is 93.2% and 90.3%, respectively.