Physical Human Activity Recognition Using Wearable Sensors

Complex & Intelligent Systems

In recent years, the adoption of machine learning has grown steadily in different fields affecting the day-to-day decisions of individuals. This paper presents an intelligent system for recognizing human’s daily activities in a complex IoT environment. An enhanced model of capsule neural network called 1D-HARCapsNe is proposed. This proposed model consists of convolution layer, primary capsule layer, activity capsules flat layer and output layer. It is validated using WISDM dataset collected via smart devices and normalized using the random-SMOTE algorithm to handle the imbalanced behavior of the dataset. The experimental results indicate the potential and strengths of the proposed 1D-HARCapsNet that achieved enhanced performance with an accuracy of 98.67%, precision of 98.66%, recall of 98.67%, and F1-measure of 0.987 which shows major performance enhancement compared to the Conventional CapsNet (accuracy 90.11%, precision 91.88%, recall 89.94%, and F1-measure 0.93).

Sensors (Basel, Switzerland), 2016

Physical activity recognition based on sensors is a growing area of interest given the great advances in wearable sensors. Applications in various domains are taking advantage of the ease of obtaining data to monitor personal activities and behavior in order to deliver proactive and personalized services. Although many activity recognition systems have been developed for more than two decades, there are still open issues to be tackled with new techniques. We address in this paper one of the main challenges of human activity recognition: Flexibility. Our goal in this work is to present artificial hydrocarbon networks as a novel flexible approach in a human activity recognition system. In order to evaluate the performance of artificial hydrocarbon networks based classifier, experimentation was designed for user-independent, and also for user-dependent case scenarios. Our results demonstrate that artificial hydrocarbon networks classifier is flexible enough to be used when building a h...

Electronics

The Human Activity Recognition (HAR) system allows various accessible entries for the early diagnosis of Diabetes as one of the nescient applications domains for the HAR. Long Short-Term Memory (LSTM) was applied and recognized 13 activities that resemble diabetes symptoms. Afterward, risk factor assessment for an experimental subject identified similar activity pattern attributes between diabetic patients and the experimental subject. Because of this, a trained LSTM model was deployed to monitor the average time length for every activity performed by the experimental subject for 30 consecutive days. Concurrently, the symptomatic diabetes activity patterns of diabetic patients were explored. The cosine similarity of activity patterns of the experimental subject and diabetic patients measured 57.39%, putting the experimental subject into moderate risk factor class. The experimental subject was clinically tested for risk factors using the diabetic clinical diagnosis process, known as ...

Sensors

Inertial sensors are a useful instrument for long term monitoring in healthcare. In many cases, inertial sensor devices can be worn as an accessory or integrated into smart textiles. In some situations, it may be beneficial to have data from multiple inertial sensors, rather than relying on a single worn sensor, since this may increase the accuracy of the analysis and better tolerate sensor errors. Integrating multiple sensors into clothing improves the feasibility and practicality of wearing multiple devices every day, in approximately the same location, with less likelihood of incorrect sensor orientation. To facilitate this, the current work investigates the consequences of attaching lightweight sensors to loose clothes. The intention of this paper is to discuss how data from these clothing sensors compare with similarly placed body worn sensors, with additional consideration of the resulting effects on activity recognition. This study compares the similarity between the two sign...

Sensors (Basel, Switzerland), 2018

Wearable inertial devices have recently been used to evaluate spatiotemporal parameters of gait in daily life situations. Given the heterogeneity of gait patterns in children with cerebral palsy (CP), the sensor placement and analysis algorithm may influence the validity of the results. This study aimed at comparing the spatiotemporal measurement performances of three wearable configurations defined by different sensor positioning on the lower limbs: (1) shanks and thighs, (2) shanks, and (3) feet. The three configurations were selected based on their potential to be used in daily life for children with CP and typically developing (TD) controls. For each configuration, dedicated gait analysis algorithms were used to detect gait events and compute spatiotemporal parameters. Fifteen children with CP and 11 TD controls were included. Accuracy, precision, and agreement of the three configurations were determined in comparison with an optoelectronic system as a reference. The three confi...

Sensors

Although many methods have been developed to detect walking by using body-worn inertial sensors, their performances decline when gait patterns become abnormal, as seen in children with cerebral palsy (CP). The aim of this study was to evaluate if fine-tuning an existing walking bouts (WB) detection algorithm by various thresholds, customized at the individual or group level, could improve WB detection in children with CP and typical development (TD). Twenty children (10 CP, 10 TD) wore 4 inertial sensors on their lower limbs during laboratory and out-laboratory assessments. Features extracted from the gyroscope signals recorded in the laboratory were used to tune thresholds of an existing walking detection algorithm for each participant (individual-based personalization: Indiv) or for each group (population-based customization: Pop). Out-of-laboratory recordings were analyzed for WB detection with three versions of the algorithm (i.e., original fixed thresholds and adapted threshold...

Sensors

Background: A nanomaterial-based electronic-skin (E-Skin) wearable sensor has been successfully used for detecting and measuring body movements such as finger movement and foot pressure. The ultrathin and highly sensitive characteristics of E-Skin sensor make it a suitable alternative for continuously out-of-hospital lumbar–pelvic movement (LPM) monitoring. Monitoring these movements can help medical experts better understand individuals’ low back pain experience. However, there is a lack of prior studies in this research area. Therefore, this paper explores the potential of E-Skin sensors to detect and measure the anatomical angles of lumbar–pelvic movements by building a linear relationship model to compare its performance to clinically validated inertial measurement unit (IMU)-based sensing system (ViMove). Methods: The paper first presents a review and classification of existing wireless sensing technologies for monitoring of body movements, and then it describes a series of exp...

IoT

Machine learning is a powerful tool for data classification and has been used to classify movement data recorded by wearable inertial sensors in general living and sports. Inertial sensors can provide valuable biofeedback in combat sports such as boxing; however, the use of such technology has not had a global uptake. If simple inertial sensor configurations can be used to automatically classify strike type, then cumbersome tasks such as video labelling can be bypassed and the foundation for automated workload monitoring of combat sport athletes is set. This investigation evaluates the classification performance of six different supervised machine learning models (tuned and untuned) when using two simple inertial sensor configurations (configuration 1—inertial sensor worn on both wrists; configuration 2—inertial sensor worn on both wrists and third thoracic vertebrae [T3]). When trained on one athlete, strike prediction accuracy was good using both configurations (sensor configurati...

Sensors

This article presents an overview of fifty-eight articles dedicated to the evaluation of physical activity in free-living conditions using wearable motion sensors. This review provides a comprehensive summary of the technical aspects linked to sensors (types, number, body positions, and technical characteristics) as well as a deep discussion on the protocols implemented in free-living conditions (environment, duration, instructions, activities, and annotation). Finally, it presents a description and a comparison of the main algorithms and processing tools used for assessing physical activity from raw signals.

Journal of applied physiology (Bethesda, Md. : 1985), 2017

Accelerometry is increasingly used to quantify physical activity (PA) and related energy expenditure (EE). Linear regression models designed to derive PAEE from accelerometry-counts have shown their limits, mostly due to the lack of consideration of the nature of activities performed. Here we tested whether a model coupling an automatic activity/posture recognition (AAR) algorithm with an activity-specific count-based model, developed in 61 subjects in laboratory conditions, improved PAEE and total EE (TEE) predictions from hip-worn triaxial-accelerometer (ActigraphGT3X+TM) in free-living conditions. Data from two independent subject groups of varying body mass index and age were considered: 20 subjects engaged in a 3h urban-circuit, with activity-by-activity reference PAEE from combined heart-rate and accelerometry monitoring (ActiheartTM); and 56 subjects involved in a 14-day trial, with PAEE and TEE measured using the doubly-labeled-water method. PAEE was estimated from accelerom...

International Journal of Distributed Sensor Networks

Human action recognition supported by highly accurate specialized systems, ambulatory systems, or wireless sensor networks has a tremendous potential in the areas of healthcare or wellbeing monitoring. Recently, several studies carried out focused on the recognition of actions using wearable inertial sensors, in which raw sensor data are used to build classification models, and in a few of them high-level representations are obtained which are directly related to anatomical characteristics of the human body. This research focuses on classifying a set of activities of daily living, such as functional mobility, and instrumental activities of daily living, such as preparing meals, performed by test subjects in their homes in naturalistic conditions. The joint angles of upper and lower limbs are estimated using information from five wearable inertial sensors placed on the body of five test subjects. A set of features related to human limb motions is extracted from the orientation signal...

Advances in Science, Technology and Engineering Systems Journal, 2020

Article history: Received: 24 May, 2020 Accepted: 19 July, 2020 Online: 09 August, 2020 Internet of Things provides a virtual view of real-life things by guiding challenges faced by persons in daily life. It is reforming our world with trillions of sensors and other IoT enabled devices by creating a smart environment. Effective use of IoT sensors in various smart IoT applications is very important. After analyzing different sensor applications, this paper presents various types of wearable sensors used for monitoring of human activities along with different locations optimal for their placement. This paper enlightens sensors suitable for any particular application. IoT has opened up a new avenue of research in the field of human activity recognition using wearable sensors. It provides an individual's valuable information regarding functional ability and lifestyle. Human activities are monitored automatically to provide personalized support to different individuals. Recently, var...

2021

Recent advancement in smart phones and computing technologies has played a vital role in people’s life. Develop a model to detect the human basic dynamic activities such as Amble, Climb stairs, coming down the stairs into the floor and human basic static activities like Sitting, Standing or Laying using the person’s smart phone and computers are the major work of this paper. Conventional Machine learning models like Logistic Regression, SVC, Decision tree, etc. results are compared with a recurrent deep neural network model named as Long Short Term Memory (LSTM). LSTM is proposed to detect the human behavior based on Human Activity Recognition (HAR) dataset. The data is monitored and recorded with the aid of sensors like accelerometer and Gyroscope in the user smart phone. HAR dataset is collected from 30 persons, performing different activities with a smart phone to their waists. The testing of the model is evaluated with respect to accuracy and efficiency. The designed activity re...

Energies

Building Energy and Comfort Management (BECM) systems have the potential to considerably reduce costs related to energy consumption and improve the efficiency of resource exploitation, by implementing strategies for resource management and control and policies for Demand-Side Management (DSM). One of the main requirements for such systems is to be able to adapt their management decisions to the users’ specific habits and preferences, even when they change over time. This feature is fundamental to prevent users’ disaffection and the gradual abandonment of the system. In this paper, a sensor-based system for analysis of user habits and early detection and prediction of user activities is presented. To improve the resulting accuracy, the system incorporates statistics related to other relevant external conditions that have been observed to be correlated (e.g., time of the day). Performance evaluation on a real use case proves that the proposed system enables early recognition of activi...

Sensors

This article presents a machine learning methodology for diagnosing Parkinson’s disease (PD) based on the use of vertical Ground Reaction Forces (vGRFs) data collected from the gait cycle. A classification engine assigns subjects to healthy or Parkinsonian classes. The diagnosis process involves four steps: data pre-processing, feature extraction and selection, data classification and performance evaluation. The selected features are used as inputs of each classifier. Feature selection is achieved through a wrapper approach established using the random forest algorithm. The proposed methodology uses both supervised classification methods including K-nearest neighbour (K-NN), decision tree (DT), random forest (RF), Naïve Bayes (NB), support vector machine (SVM) and unsupervised classification methods such as K-means and the Gaussian mixture model (GMM). To evaluate the effectiveness of the proposed methodology, an online dataset collected within three different studies is used. This ...

2017

In this paper, we propose an algorithmic approach for a motion analysis framework to automatically recognize local muscular endurance (LME) exercises and to count their repetitions using a wrist-worn inertial sensor. LME exercises are prescribed for cardiovascular disease rehabilitation. As a technical solution, we propose activity recognition based on machine learning. We developed an algorithm to automatically segment the captured data from all participants. Relevant time and frequency domain features were extracted using a sliding window technique. Principal component analysis (PCA) was applied for dimensionality reduction of the extracted features. We trained 15 binary classifiers using support vector machine (SVM) to recognize individual LME exercises, achieving overall accuracy of more than 98%. We applied grid search technique to obtain the optimal SVM hyperplane parameters. The learning curves (mean ± stdev) for each model is investigated to verify that the models were not o...

Sensors

This study aimed to develop a wearable sensor system, using machine-learning models, capable of accurately estimating peak ground reaction force (GRF) during ballet jumps in the field. Female dancers (n = 30) performed a series of bilateral and unilateral ballet jumps. Dancers wore six ActiGraph Link wearable sensors (100 Hz). Data were collected simultaneously from two AMTI force platforms and synchronised with the ActiGraph data. Due to sensor hardware malfunctions and synchronisation issues, a multistage approach to model development, using a reduced data set, was taken. Using data from the 14 dancers with complete multi-sensor synchronised data, the best single sensor was determined. Subsequently, the best single sensor model was refined and validated using all available data for that sensor (23 dancers). Root mean square error (RMSE) in body weight (BW) and correlation coefficients (r) were used to assess the GRF profile, and Bland–Altman plots were used to assess model peak GR...

Proceedings of the 16th ACM Conference on Embedded Networked Sensor Systems, 2018

Smart apparel with embedded sensors have the potential to revolutionize human behavior sensing by leveraging everyday clothing as the sensing substrate. However, existing textile-based sensing techniques rely on tight-fitting garments to obtain sufficient signal to noise, making it uncomfortable to wear and limiting the technology to niche applications like athletic performance monitoring. Our solution leverages functionalized fabric to measure the triboelectric charges induced by folding and compression of the textile itself, making it a more natural fit for everyday clothing. However, the large sensing surface of a functionalized textile also increases body-coupled noise and motion artifacts, and introduces new challenges in how we suppress noise to detect the weak triboelectric signal. We address these challenges using a combination of textile, electronics, and signal analysis-based innovations, and robustly sense joint motions by improving SNR and extracting highly discriminativ...

Journal of Rehabilitation and Assistive Technologies Engineering, 2020

Introduction Falls cause major expenses in the healthcare sector. We investigate the ability of supporting a fall risk assessment by introducing algorithms for automated assessments of standardized fall risk-related tests via wearable devices. Methods In a study, 13 participants conducted the standardized 6-Minutes Walk Test, the Timed-Up-and-Go Test, the 30-Second Sit-to-Stand Test, and the 4-Stage Balance Test repeatedly, producing 226 tests in total. Automated algorithms computed by wearable devices, as well as a visual analysis of the recorded data streams, were compared to the observational results conducted by physiotherapists. Results There was a high congruence between automated assessments and the ground truth for all four test types (ranging from 78.15% to 96.55%), with deviations ranging all well within one standard deviation of the ground truth. Fall risk (assessed by questionnaire) correlated with the individual tests. Conclusions The automated fall risk assessment usin...

Sensors

The field of human activity recognition (HAR) often utilizes wearable sensors and machine learning techniques in order to identify the actions of the subject. This paper considers the activity recognition of walking and running while using a support vector machine (SVM) that was trained on principal components derived from wearable sensor data. An ablation analysis is performed in order to select the subset of sensors that yield the highest classification accuracy. The paper also compares principal components across trials to inform the similarity of the trials. Five subjects were instructed to perform standing, walking, running, and sprinting on a self-paced treadmill, and the data were recorded while using surface electromyography sensors (sEMGs), inertial measurement units (IMUs), and force plates. When all of the sensors were included, the SVM had over 90% classification accuracy using only the first three principal components of the data with the classes of stand, walk, and run...

Sustainability

SARS-CoV-2, a tiny virus, is severely affecting the social, economic, and environmental sustainability of our planet, causing infections and deaths (2,674,151 deaths, as of 17 March 2021), relationship breakdowns, depression, economic downturn, riots, and much more. The lessons that have been learned from good practices by various countries include containing the virus rapidly; enforcing containment measures; growing COVID-19 testing capability; discovering cures; providing stimulus packages to the affected; easing monetary policies; developing new pandemic-related industries; support plans for controlling unemployment; and overcoming inequalities. Coordination and multi-term planning have been found to be the key among the successful national and global endeavors to fight the pandemic. The current research and practice have mainly focused on specific aspects of COVID-19 response. There is a need to automate the learning process such that we can learn from good and bad practices dur...

Sensors, 2017

In addition to routine clinical examination, unobtrusive and physical monitoring of Rheumatoid Arthritis (RA) patients provides an important source of information to enable understanding the impact of the disease on quality of life. Besides an increase in sedentary behaviour, pain in RA can negatively impact simple physical activities such as getting out of bed and standing up from a chair. The objective of this work is to develop a method that can generate fine-grained actigraphies to capture the impact of the disease on the daily activities of patients. A processing methodology is presented to automatically tag activity accelerometer data from a cohort of moderate-to-severe RA patients. A study of procesing methods based on machine learning and deep learning is provided. Thirty subjects, 10 RA patients and 20 healthy control subjects, were recruited in the study. A single tri-axial accelerometer was attached to the position of the fifth lumbar vertebra (L5) of each subject with a tag prediction granularity of 3 s. The proposed method is capable of handling unbalanced datasets from tagged data while accounting for long-duration activities such as sitting and lying, as well as short transitions such as sit-to-stand or lying-to-sit. The methodology also includes a novel mechanism for automatically applying a threshold to predictions by their confidence levels, in addition to a logical filter to correct for infeasible sequences of activities. Performance tests showed that the method was able to achieve around 95% accuracy and 81% F-score. The produced actigraphies can be helpful to generate objective RA disease-specific markers of patient mobility in-between clinical site visits.

Sensors, 2016

Human activity recognition has gained more interest in several research communities given that understanding user activities and behavior helps to deliver proactive and personalized services. There are many examples of health systems improved by human activity recognition. Nevertheless, the human activity recognition classification process is not an easy task. Different types of noise in wearable sensors data frequently hamper the human activity recognition classification process. In order to develop a successful activity recognition system, it is necessary to use stable and robust machine learning techniques capable of dealing with noisy data. In this paper, we presented the artificial hydrocarbon networks (AHN) technique to the human activity recognition community. Our artificial hydrocarbon networks novel approach is suitable for physical activity recognition, noise tolerance of corrupted data sensors and robust in terms of different issues on data sensors. We proved that the AHN classifier is very competitive for physical activity recognition and is very robust in comparison with other well-known machine learning methods.

Sensors, 2020

Human activity recognition (HAR) is growing in popularity due to its wide-ranging applications in patient rehabilitation and movement disorders. HAR approaches typically start with collecting sensor data for the activities under consideration and then develop algorithms using the dataset. As such, the success of algorithms for HAR depends on the availability and quality of datasets. Most of the existing work on HAR uses data from inertial sensors on wearable devices or smartphones to design HAR algorithms. However, inertial sensors exhibit high noise that makes it difficult to segment the data and classify the activities. Furthermore, existing approaches typically do not make their data available publicly, which makes it difficult or impossible to obtain comparisons of HAR approaches. To address these issues, we present wearable HAR (w-HAR) which contains labeled data of seven activities from 22 users. Our dataset’s unique aspect is the integration of data from inertial and wearable...

Sustainability, 2021

Sustainable work aims at improving working conditions to allow workers to effectively extend their working life. In this context, occupational safety and well-being are major concerns, especially in labor-intensive fields, such as construction-related work. Internet of Things and wearable sensors provide for unobtrusive technology that could enhance safety using human activity recognition techniques, and has the potential of improving work conditions and health. However, the research community lacks commonly used standard datasets that provide for realistic and variating activities from multiple users. In this article, our contributions are threefold. First, we present VTT-ConIoT, a new publicly available dataset for the evaluation of HAR from inertial sensors in professional construction settings. The dataset, which contains data from 13 users and 16 different activities, is collected from three different wearable sensor locations.Second, we provide a benchmark baseline for human a...

PLOS ONE, 2019

Recent increases in the use of and applications for wearable technology has opened up many new avenues of research. In this paper, we consider the use of lifelogging and GPS data to extend fine-grained movement analysis for improving applications in health and safety. We first design a framework to solve the problem of indoor and outdoor movement detection from sensor readings associated with images captured by a lifelogging wearable device. Second we propose a set of measures related with hazard on the road network derived from the combination of GPS movement data, road network data and the sensor readings from a wearable device. Third, we identify the relationship between different sociodemographic groups and the patterns of indoor physical activity and sedentary behaviour routines as well as disturbance levels on different road settings.

Sensors, 2021

Human activity recognition (HAR) aims to recognize the actions of the human body through a series of observations and environmental conditions. The analysis of human activities has drawn the attention of the research community in the last two decades due to its widespread applications, diverse nature of activities, and recording infrastructure. Lately, one of the most challenging applications in this framework is to recognize the human body actions using unobtrusive wearable motion sensors. Since the human activities of daily life (e.g., cooking, eating) comprises several repetitive and circumstantial short sequences of actions (e.g., moving arm), it is quite difficult to directly use the sensory data for recognition because the multiple sequences of the same activity data may have large diversity. However, a similarity can be observed in the temporal occurrence of the atomic actions. Therefore, this paper presents a two-level hierarchical method to recognize human activities using ...

Sensors, 2021

Human activity recognition (HAR) by wearable sensor devices embedded in the Internet of things (IOT) can play a significant role in remote health monitoring and emergency notification to provide healthcare of higher standards. The purpose of this study is to investigate a human activity recognition method of accrued decision accuracy and speed of execution to be applicable in healthcare. This method classifies wearable sensor acceleration time series data of human movement using an efficient classifier combination of feature engineering-based and feature learning-based data representation. Leave-one-subject-out cross-validation of the method with data acquired from 44 subjects wearing a single waist-worn accelerometer on a smart textile, and engaged in a variety of 10 activities, yielded an average recognition rate of 90%, performing significantly better than individual classifiers. The method easily accommodates functional and computational parallelization to bring execution time s...

SSRN Electronic Journal, 2018

Pattern recognition is an interesting research area, especially for human activity recognition (HAR). The purpose of HAR study is to automatically detect human activities from the information acquired from different sensors using statistical techniques. Despite the great efforts for HAR is still a challenging task because of high dimensionality problems. Several approaches have recently been developed to solve these problems. The mean objective of this work is to deal with high dimensionality in HAR modeling to minimize the number of features and maximize the classification accuracy.

International Journal of Distributed Sensor Networks, 2020

We propose to perform wearable sensors-based human physical activity recognition. This is further extended to an Internet-of-Things (IoT) platform which is based on a web-based application that integrates wearable sensors, smartphones, and activity recognition. To this end, a smartphone collects the data from wearable sensors and sends it to the server for processing and recognition of the physical activity. We collect a novel data set of 13 physical activities performed both indoor and outdoor. The participants are from both the genders where their number per activity varies. During these activities, the wearable sensors measure various body parameters via accelerometers, gyroscope, magnetometers, pressure, and temperature. These measurements and their statistical are then represented in features vectors that used to train and test supervised machine learning algorithms (classifiers) for activity recognition. On the given data set, we evaluate a number of widely known classifiers s...

Innovative Mobile and Internet Services in Ubiquitous Computing, 2018

Acoustic Scene Classification (ASC) and Sound Event Detection (SED) are two separate tasks in the field of computational sound scene analysis. In this work, we present a new dataset with both sound scene and sound event labels and use this to demonstrate a novel method for jointly classifying sound scenes and recognizing sound events. We show that by taking a joint approach, learning is more efficient and whilst improvements are still needed for sound event detection, SED results are robust in a dataset where the sample distribution is skewed towards sound scenes.

Smart Computing and Intelligence

Improvements during portable and electronic medical services are changing a contribution on behalf of together specialists along with patients during an advanced medical care structure through increasing the capacities of physiological observing gadgets. Lately there is fast exploration going in remote sensors for wireless Healthcare. These remote sensors give different functionalities in universal climate that permits sensors to associate with web anytime and anyplace. Wearable sensor frameworks which take into consideration distant or self-checking of wellbeing related boundaries are viewed as one intends to ease the results of segment change. Sensors assume a significant function in medical care industry. In spite of huge advancement inside the inspection tool industry, the broad reconciliation of this innovation into clinical practice stays restricted. Then again Social media has pushed forward of simply being an apparatus for young adult clients to share their own lives (data, life info) however has additionally been taken up by medical care partners to scatter health data around the world. The function of web-based media in medical services industry and effect of patient commitment has moved to focused phase and is constantly moved by quiet interest, mobile innovation, and mounting impact of the local electronic age. The movement with which the social action is developing, and it is essential to predict what affect these electronic stages are having on the recovery and result of medical care intercessions. The motivation behind this review is to sum up the turns of events and clinical utility of brilliant wearable body sensors. A review taken how far online media has been end up being powerful in spreading wellbeing data and how much advance methodologies have been useful

Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies

The recent prevalence of machine learning-based techniques and smart device embedded sensors has enabled widespread human-centric sensing applications. However, these applications are vulnerable to false data injection attacks (FDIA) that alter a portion of the victim's sensory signal with forged data comprising a targeted trait. Such a mixture of forged and valid signals successfully deceives the continuous authentication system (CAS) to accept it as an authentic signal. Simultaneously, introducing a targeted trait in the signal misleads human-centric applications to generate specific targeted inference; that may cause adverse outcomes. This paper evaluates the FDIA's deception efficacy on sensor-based authentication and human-centric sensing applications simultaneously using two modalities - accelerometer, blood volume pulse signals. We identify variations of the FDIA such as different forged signal ratios, smoothed and non-smoothed attack samples. Notably, we present a no...

Advances in Intelligent Systems and Computing, 2019

This paper presents an equipment that can be used in the medical recovery activity, including for people with special needs. The equipment is composed of two human body motion tracking systems. The first system is based on video tracking technology and uses a Kinect sensor. The data captured by this video sensor will be used to animate an avatar from a dedicated software application, after which the patient can analyze the accuracy of real-time movements on a display. The second tracking system uses inertial sensors mounted on the patient's lower limbs. These sensors return Euler angles that are processed, recorded and used for later evaluate. For this analysis, a human body model, using non-deformable segments and joints, able to reproduce human body movements, was developed. This skeleton model can be superimposed on the CATIA software mannequin and can be animated using a macro based on the Euler angles read by the sensors so that it can analyze the patient's movements using data records gathered during recovery session.

Biosensors

Many studies have explored divergent deep neural networks in human activity recognition (HAR) using a single accelerometer sensor. Multiple types of deep neural networks, such as convolutional neural networks (CNN), long short-term memory (LSTM), or their hybridization (CNN-LSTM), have been implemented. However, the sensor orientation problem poses challenges in HAR, and the length of windows as inputs for the deep neural networks has mostly been adopted arbitrarily. This paper explores the effect of window lengths with orientation invariant heuristic features on the performance of 1D-CNN-LSTM in recognizing six human activities; sitting, lying, walking and running at three different speeds using data from an accelerometer sensor encapsulated into a smartphone. Forty-two participants performed the six mentioned activities by keeping smartphones in their pants pockets with arbitrary orientation. We conducted an inter-participant evaluation using 1D-CNN-LSTM architecture. We found tha...

Computational Intelligence and Neuroscience

With the advent of the Internet of Things (IoT), human-assistive technologies in healthcare services have reached the peak of their application in terms of diagnosis and treatment process. These devices must be aware of human movements to provide better aid in clinical applications as well as the user’s daily activities. In this context, real-time gait analysis remains to be key catalyst for developing intelligent assistive devices. In addition to machine and deep learning algorithms, gait recognition systems have significantly improved in terms of high accuracy recognition. However, most of the existing models are focused on improving gait recognition while ignoring the computational overhead that affects the accuracy of detection and even remains unsuitable for real-time implementation. In this research paper, we proposed a hybrid gated recurrent unit (GRU) based on BAT-inspired extreme convolutional networks (BAT-ECN) for the effective recognition of human activities using gait d...

International Journal of Information Technology, 2021

Identification of human physical activities is an active research area since long due to its application in personalized health and fitness monitoring. The performance accuracy of human activity recognition (HAR) models mainly depend on the features which are extracted from domain knowledge. The features are the input of the classification algorithm to efficiently identify human physical activities. Manually extracted features (handcrafted) need expert domain knowledge. Thus these features have significant importance to identify different human activities. Recently deep learning methods are utilized to extract the features automatically from raw sensory data for HAR models. However, state-of-the-art HAR literature established that the importance of handcrafted features can't be ignored as it is extracted from expert domain knowledge. Thus, in this paper we use the fusion of both the handcrafted features and automatically extracted features using deep learning (DL) for HAR model to enhance the performance of HAR. Extensive experimental results demonstrate that our proposed feature fusion based HAR model gives higher accuracy compared with state-ofthe-art HAR literature for both the self collected and public dataset.

International Journal of Social Robotics

Socially Assistive Robots (SARs) are designed to support us in our daily life as a companion, and assistance but also to support the caregivers’ work. SARs should show personalized and human-like behavior to improve their acceptance and, consequently, their use. Additionally, they should be trustworthy by caregivers and professionals to be used as support for their work (e.g. objective assessment, decision support tools). In this context the aim of the paper is dual. Firstly, this paper aims to present and discuss the robot behavioral model based on sensing, perception, decision support, and interaction modules. The novel idea behind the proposed model is to extract and use the same multimodal features set for two purposes: (i) to profile the user, so to be used by the caregiver as a decision support tool for the assessment and monitoring of the patient; (ii) to fine-tune the human–robot interaction if they can be correlated to the social cues. Secondly, this paper aims to test in a...

International Journal of Medical Informatics, 2021

Background and Objective: The assessment of dependence in older adults currently requires a manual collection of data taken from questionnaires. This process is time consuming for the clinicians and intrudes the daily life of the elderly. This paper aims to semi-automate the acquisition and analysis of health data to assess and predict the dependence in older adults while executing one instrumental activity of daily living (IADL). Methods: In a mobile-health (m-health) scenario, we analyze whether the acquisition of data through wearables during the performance of IADLs, and with the help of machine learning techniques could replace the traditional questionnaires to evaluate dependence. To that end, we collected data from wearables, while older adults do the shopping activity. A trial supervisor (TS) labelled the different shopping stages (SS) in the collected data. We performed data pre-processing techniques over those SS and analyzed them with three machine learning algorithms: k-Nearest Neighbors (k-NN), Random Forest (RF) and Support Vector Machines (SVM). Results: Our results confirm that it is possible to replace the traditional questionnaires with wearable data. In particular, the best learning algorithm we tried reported an accuracy of 97% in the assessment of dependence. We tuned the hyperparameters of this algorithm and used embedded feature selection technique to get the best performance with a subset of only 10 features out of the initial 85. This model considers only features extracted from four sensors of a single wearable: accelerometer, heart rate, electrodermal activity and temperature. Although these features are not observational, our current proposal is semi-automatic, because it needs a TS labelling the SS (with a smartphone application). In the future, this labelling process could be automatic as well. Conclusions: Our method can semi-automatically assess the dependence, without disturbing daily activities of elderly people. This method can save clinicians' time in the evaluation of dependence in older adults and reduce healthcare costs.

Applied Sciences

This paper presents a comparison of mathematical and cinematic motion analysis regarding the accuracy of the detection of alterations in the patterns of positional sequence during biceps-curl lifting exercise. Two different methods, one with and one without metric data from the environment, were used to identify the changes. Ten volunteers performed a standing biceps-curl exercise with additional loads. A smartphone recorded their movements in the sagittal plane, providing information on joints and barbell sequential position changes during each lift attempt. An analysis of variance revealed significant differences in joint position (p < 0.05) among executions with three different loads. Hidden Markov models were trained with data from the bi-dimensional coordinates of the joint positional sequence to identify meaningful alteration with load increment. Tests of agreement tests between the results provided by the models with the environmental measurements, as well as those from im...

Scientific Reports

Inertial Measurement Units (IMUs) are a potential way to monitor the mobility of people outside clinical or laboratory settings at an acceptable cost. To increase accuracy, multiple IMUs can be used. By embedding multiple sensors into everyday clothing, it is possible to simplify having to put on individual sensors, ensuring sensors are correctly located and oriented. This research demonstrates how clothing-mounted IMU readings can be used to identify 4 common postures: standing, sitting, lying down and sitting on the floor. Data were collected from 5 healthy adults, with each providing 1–4 days of data with approximately 5 h each day. Each day, participants performed a fixed set of activities that were video-recorded to provide a ground truth. This is an analysis of accelerometry data from 3 sensors incorporated into right trouser-leg at the waist, thigh and ankle. Data were classified as static/ dynamic activities using a K-nearest neighbour (KNN) algorithm. For static activities,...

Wireless Personal Communications, 2018

Medical sensor nodes are used in pervasive healthcare applications like remote patient monitoring, elderly care to collect patients vital signs for identifying medical emergency. These resource restricted sensor nodes are prone to various malicious attacks, data faults and data losses. Presence of faulty data, data loss in collected patient data may lead to incorrect analysis of patient condition, which decreases the reliability of pervasive healthcare system. The aim of this work is to alert the caregiver and raise the alarm only when the patient enters into medical emergency situation. The proposed scheme also reduces the false alarms and alerts caused by data fault and misbehaving sensor nodes. To achieve this, we introduce a context aware trust management scheme for data fault detection, data reconstruction and event detection in pervasive healthcare systems. It employs heuristic functions, data correlation and contextual information based algorithms to identify the data faults and events. It also reconstructs the data faults and data loss for identifying patient condition. Performance of this approach is evaluated with the help of real data samples collected by medical sensor network prototype of remote patient monitoring application. The experimental results show that the proposed trust scheme outperforms state-of-the-art techniques and achieves good detection accuracy in data fault detection and event detection.

Buildings, 2020

The digitalization of the construction industry (CI) has the aim—among others—to raise the bar of overall productivity. The craft workforce is very relevant on the overall value-chain. Therefore, a boost in this dimension impacts the entire sector. There is a gap in proper methodologies to measure and model productivity. Construction 4.0 novelties provide new approaches for its evaluation and progress. This communication presents a review of workforce productivity assessment and delivers methods focusing primarily on craft workers motion monitoring. Products and services opportunities from Construction 4.0 in the spectrum of craft workforce management include support by embedded sensors for data collection that allow near real-time monitoring. The work developed led to the systematization of a framework to standardize craft workers’ motion productivity. The craft workforce motion productivity framework, Worker 4.0, tenders nine processes integrated on a flowchart to streamline task ...

Signals, 2021

Fall events in elderly populations often result in serious injury and may lead to long-term disability and/or death. While many fall detection systems have been developed using wearable sensors to distinguish falls from other daily activities, detection sensitivity and specificity decreases when exposed to more rigorous activities such as running and jumping. This research uses time-frequency analysis of accelerometer-only activity data to develop a strategy for improving fall detection accuracy. In this study, a wireless sensor system (WSS) consisting of a three-axis accelerometer, microprocessor and wireless communications module is used to collect daily activities performed following a script in the laboratory setting. Experiments were conducted on 36 healthy human subjects performing four types of falls (i.e., forward, backward, and left/right sideway falls) as well as normal movements such as standing, walking, stand-to-sit, sit-to-stand, stepping, running and jumping. In total...

2017 IEEE 36th International Performance Computing and Communications Conference (IPCCC)

Most of the studies on human activity recognition using smartphones and smartwatches are performed in an offline manner. In such studies, collected data is analyzed in machine learning tools with less focus on the resource consumption of these devices for running an activity recognition system. In this paper, we analyze the resource consumption of human activity recognition on both smartphones and smartwatches, considering six different classifiers, three different sensors, different sampling rates and window sizes. We study the CPU, memory and battery usage with different parameters, where the smartphone is used to recognize seven physical activities and the smartwatch is used to recognize smoking activity. As a result of this analysis, we report that classification function takes a very small amount of CPU time out of total app's CPU time while sensing and feature calculation consume most of it. When an additional sensor is used besides an accelerometer, such as gyroscope, CPU usage increases significantly. Analysis results also show that increasing the window size reduces the resource consumption more than reducing the sampling rate. As a final remark, we observe that a more complex model using only the accelerometer is a better option than using a simple model with both accelerometer and gyroscope when resource usage is to be reduced.

Sensors, 2018

Human body motion analysis based on wearable inertial measurement units (IMUs) receives a lot of attention from both the research community and the and industrial community. This is due to the significant role in, for instance, mobile health systems, sports and human computer interaction. In sensor based activity recognition, one of the major issues for obtaining reliable results is the sensor placement/assignment on the body. For inertial motion capture (joint kinematics estimation) and analysis, the IMU-to-segment (I2S) assignment and alignment are central issues to obtain biomechanical joint angles. Existing approaches for I2S assignment usually rely on hand crafted features and shallow classification approaches (e.g., support vector machines), with no agreement regarding the most suitable features for the assignment task. Moreover, estimating the complete orientation alignment of an IMU relative to the segment it is attached to using a machine learning approach has not been shown in literature so far. This is likely due to the high amount of training data that have to be recorded to suitably represent possible IMU alignment variations. In this work, we propose online approaches for solving the assignment and alignment tasks for an arbitrary amount of IMUs with respect to a biomechanical lower body model using a deep learning architecture and windows of 128 gyroscope and accelerometer data samples. For this, we combine convolutional neural networks (CNNs) for local filter learning with long-short-term memory (LSTM) recurrent networks as well as generalized recurrent units (GRUs) for learning time dynamic features. The assignment task is casted as a classification problem, while the alignment task is casted as a regression problem. In this framework, we demonstrate the feasibility of augmenting a limited amount of real IMU training data with simulated alignment variations and IMU data for improving the recognition/estimation accuracies. With the proposed approaches and final models we achieved 98.57% average accuracy over all segments for the I2S assignment task (100% when excluding left/right switches) and an average median angle error over all segments and axes of 2.91°f or the I2S alignment task.

Sensors, 2020

Nowadays, wearable technology can enhance physical human life-log routines by shifting goals from merely counting steps to tackling significant healthcare challenges. Such wearable technology modules have presented opportunities to acquire important information about human activities in real-life environments. The purpose of this paper is to report on recent developments and to project future advances regarding wearable sensor systems for the sustainable monitoring and recording of human life-logs. On the basis of this survey, we propose a model that is designed to retrieve better information during physical activities in indoor and outdoor environments in order to improve the quality of life and to reduce risks. This model uses a fusion of both statistical and non-statistical features for the recognition of different activity patterns using wearable inertial sensors, i.e., triaxial accelerometers, gyroscopes and magnetometers. These features include signal magnitude, positive/negat...

Natural and Applied Sciences Journal, 2019

In early 1990s, microfluidics technology was mainly aiming at the manipulation of fluids in micro-scale and nanoscale. At present, with the development of microfluidics, it has been widely used in the life science and medical researches with significant achievements. The microfluidics technology can be used in single cell capture, cell screening, and synthesis of biomacromolecules. Some microfluidic chips have already been commercialized and applied in disease detection, drug delivery and bioscience. However, the physical index oriented wearable technology ignored another part of the most important indications in health monitoring i.e. the body fluid. The body fluid in this review refers to the blood, sweat, interstitial fluid, saliva, tears, and urine. The current medical procedures for the testing of body fluid involve using highly sophisticated instrument such as atomic absorption spectrometry, ion chromatography and gas chromatograph for the detection of specific targets in body...

Applied Sciences, 2021

Human Activity Recognition (HAR) has become an active field of research in the computer vision community. Recognizing the basic activities of human beings with the help of computers and mobile sensors can be beneficial for numerous real-life applications. The main objective of this paper is to recognize six basic human activities, viz., jogging, sitting, standing, walking and whether a person is going upstairs or downstairs. This paper focuses on predicting the activities using a deep learning technique called Convolutional Neural Network (CNN) and the accelerometer present in smartphones. Furthermore, the methodology proposed in this paper focuses on grouping the data in the form of nodes and dividing the nodes into three major layers of the CNN after which the outcome is predicted in the output layer. This work also supports the evaluation of testing and training of the two-dimensional CNN model. Finally, it was observed that the model was able to give a good prediction of the act...

Electronics, 2022

Detecting stress when performing physical activities is an interesting field that has received relatively little research interest to date. In this paper, we took a first step towards redressing this, through a comprehensive review and the design of a low-cost body area network (BAN) made of a set of wearables that allow physiological signals and human movements to be captured simultaneously. We used four different wearables: OpenBCI and three other open-hardware custom-made designs that communicate via bluetooth low energy (BLE) to an external computer—following the edge-computingconcept—hosting applications for data synchronization and storage. We obtained a large number of physiological signals (electroencephalography (EEG), electrocardiography (ECG), breathing rate (BR), electrodermal activity (EDA), and skin temperature (ST)) with which we analyzed internal states in general, but with a focus on stress. The findings show the reliability and feasibility of the proposed body area...

Applied Sciences

As society moves towards a preventative approach to healthcare, there is growing interest in scientific research involving technology that can monitor and prevent adverse health outcomes. The primary objective of this paper is to develop an Internet of Things (IoT) wearable system based on Fried’s phenotype that is capable of detecting frailty. To determine user requirements, the system’s architecture was designed based on the findings of a questionnaire administered to individuals confirmed to be frail. A functional prototype was successfully developed and tested under real-world conditions. This paper introduces the methodology that was used to analyze the data collected from the prototype. It proposes an interdisciplinary approach to interpret wearable sensor data, providing a comprehensive overview through both visual representations and computational analyses facilitated by machine learning models. The findings of these analyses offer insights into the ways in which different t...

Frontiers in Medicine

Risk-taking behaviors of adult bedridden patients in neurosurgery are frequent, however little analyzed. We aimed to estimate from the literature and our clinical experience the incidence of the different clinical pictures. Risk-taking behaviors seem to be more frequent than reported. They are often minor, but they can lead to death, irrespective of the prescription of physical or chemical constraints. We also aimed to contextualize the risks, and to describe the means reducing the consequences for the patients. Two main conditions were identified, the loss of awareness of risk-taking behaviors by the patient, and uncontrolled body motions. Besides, current experience feedback analyses and new non-exclusive technological solutions could limit the complications, while improving prevention with wearable systems, neighborhood sensors, or room monitoring and service robots. Further research is mandatory to develop efficient and reliable systems avoiding complications and saving lives. E...

Sensors

Inertial Measurement Units (IMUs) have become a popular solution for tracking human motion. The main problem of using IMU data for deriving the position of different body segments throughout time is related to the accumulation of the errors in the inertial data. The solution to this problem is necessary to improve the use of IMUs for position tracking. In this work, we present several Machine Learning (ML) methods to improve the position tracking of various body segments when performing different movements. Firstly, classifiers were used to identify the periods in which the IMUs were stopped (zero-velocity detection). The models Random Forest, Support Vector Machine (SVM) and neural networks based on Long-Short-Term Memory (LSTM) layers were capable of identifying those periods independently of the motion and body segment with a substantially higher performance than the traditional fixed-threshold zero-velocity detectors. Afterwards, these techniques were combined with ML regression...

Neural Computing and Applications

Human activity recognition (HAR) using inertial motion data has gained a lot of momentum in recent years both in research and industrial applications. From the abstract perspective, this has been driven by the rapid dynamics for building intelligent, smart environments, and ubiquitous systems that cover all aspects of human life including healthcare, sports, manufacturing, commerce, etc., which necessitate and subsume activity recognition aiming at recognizing the actions, characteristics, and goals of one or more agent(s) from a temporal series of observations streamed from one or more sensors. From a more concrete and seemingly orthogonal perspective, such momentum has been driven by the ubiquity of inertial motion sensors on-board mobile and wearable devices including smartphones, smartwatches, etc. In this paper we give an introductory and a comprehensive survey to the subject from a given perspective. We focus on a subset of topics, that we think are major, that will have signi...

Journal of Intelligent & Fuzzy Systems, 2019

Accurate activity recognition plays a major role in smart homes to provide assistance and support for users, especially elderly and cognitively impaired people. To realize this task, knowledge-driven approaches are one of the emerging research areas that have shown interesting advantages and features. However, several limitations have been associated with these approaches. The produced models are usually incomplete to capture all types of human activities. This resulted in the limited ability to accurately infer users' activities. This paper presents an alternative approach by combining knowledge-driven with data-driven reasoning to allow activity models to evolve and adapt automatically based on users' particularities. Firstly, a knowledge-driven reasoning is presented for inferring an initial activity model. The model is then trained using data-driven techniques to produce a dynamic activity model that learns users' varying action. This approach has been evaluated using a publicly available dataset and the experimental results show the learned activity model yields significantly higher recognition rates compared to the initial activity model.

Journal of NeuroEngineering and Rehabilitation

Background Many patients with neurological movement disorders fear to fall while performing postural transitions without assistance, which prevents them from participating in daily life. To overcome this limitation, multi-directional Body Weight Support (BWS) systems have been developed allowing them to perform training in a safe environment. In addition to overground walking, these innovative/novel systems can assist patients to train many more gait-related tasks needed for daily life under very realistic conditions. The necessary assistance during the users’ movements can be provided via task-dependent support designs. One remaining challenge is the manual switching between task-dependent supports. It is error-prone, cumbersome, distracts therapists and patients, and interrupts the training workflow. Hence, we propose a real-time motion onset recognition model that performs automatic support switching between standing-up and sitting-down transitions and other gait-related tasks (8...

IEEE Access

, from funds for the implementation of KITT4SME (platform-enabled KITs of artificial intelligence for an easy uptake by SMEs) Project, through the European Commission Horizon 2020 (H2020) Program, under GA 952119 (https://kitt4sme.eu/); and in part by the Warsaw University of Technology for the Article processing Charge (APC).

PeerJ, 2022

Deep learning (DL) models are very useful for human activity recognition (HAR); these methods present better accuracy for HAR when compared to traditional, among other advantages. DL learns from unlabeled data and extracts features from raw data, as for the case of time-series acceleration. Sliding windows is a feature extraction technique. When used for preprocessing time-series data, it provides an improvement in accuracy, latency, and cost of processing. The time and cost of preprocessing can be beneficial especially if the window size is small, but how small can this window be to keep good accuracy? The objective of this research was to analyze the performance of four DL models: a simple deep neural network (DNN); a convolutional neural network (CNN); a long short-term memory network (LSTM); and a hybrid model (CNN-LSTM), when variating the sliding window size using fixed overlapped windows to identify an optimal window size for HAR. We compare the effects in two acceleration sources': wearable inertial measurement unit sensors (IMU) and motion caption systems (MOCAP). Moreover, short sliding windows of sizes 5, 10, 15, 20, and 25 frames to long ones of sizes 50, 75, 100, and 200 frames were compared. The models were fed using raw acceleration data acquired in experimental conditions for three activities: walking, sitto-stand, and squatting. Results show that the most optimal window is from 20-25 frames (0.20-0.25s) for both sources, providing an accuracy of 99,07% and F1-score of 87,08% in the (CNN-LSTM) using the wearable sensors data, and accuracy of 98,8% and F1-score of 82,80% using MOCAP data; similar accurate results were obtained with the LSTM model. There is almost no difference in accuracy in larger frames (100, 200). However, smaller windows present a decrease in the F1-score. In regard to inference time, data with a sliding window of 20 frames can be preprocessed around 4x (LSTM) and 2x (CNN-LSTM) times faster than data using 100 frames. How to cite this article Jaén-Vargas M, Reyes Leiva KM, Fernandes F, Barroso Gonçalves S, Tavares Silva M, Lopes DM, Serrano Olmedo JJ. 2022. Effects of sliding window variation in the performance of acceleration-based human activity recognition using deep learning models.

Sensors

By observing the actions taken by operators, it is possible to determine the risk level of a work task. One method for achieving this is the recognition of human activity using biosignals and inertial measurements provided to a machine learning algorithm performing such recognition. The aim of this research is to propose a method to automatically recognize physical exertion and reduce noise as much as possible towards the automation of the Job Strain Index (JSI) assessment by using a motion capture wearable device (MindRove armband) and training a quadratic support vector machine (QSVM) model, which is responsible for predicting the exertion depending on the patterns identified. The highest accuracy of the QSVM model was 95.7%, which was achieved by filtering the data, removing outliers and offsets, and performing zero calibration; in addition, EMG signals were normalized. It was determined that, given the job strain index’s purpose, physical exertion detection is crucial to computi...

Sensors, 2020

The population in developed countries is aging and this fact results in high elderly health costs, as well as a decrease in the number of active working members to support these costs. This could lead to a collapse of the current systems. One of the first insights of the decline in elderly people is frailty, which could be decelerated if it is detected at an early stage. Nowadays, health professionals measure frailty manually through questionnaires and tests of strength or gait focused on the physical dimension. Sensors are increasingly used to measure and monitor different e-health indicators while the user is performing Basic Activities of Daily Life (BADL). In this paper, we present a system based on microservices architecture, which collects sensory data while the older adults perform Instrumental ADLs (IADLs) in combination with BADLs. IADLs involve physical dimension, but also cognitive and social dimensions. With the sensory data we built a machine learning model to assess frailty status which outperforms the previous works that only used BADLs. Our model is accurate, ecological, non-intrusive, flexible and can help health professionals to automatically detect frailty.

Journal of NeuroEngineering and Rehabilitation

Background To evaluate the relationship between measures of neuromuscular impairment and limb accelerations (LA) collected during sleep among individuals with chronic spinal cord injury (SCI) to provide evidence of construct and concurrent validity for LA as a clinically meaningful measure. Methods The strength (lower extremity motor score), sensation (summed lower limb light touch scores), and spasticity (categorized lower limb Modified Ashworth Scale) were measured from 40 adults with chronic (≥ 1 year) SCI. Demographics, pain, sleep quality, and other covariate or confounding factors were measured using self-report questionnaires. Each participant then wore ActiGraph GT9X Link accelerometers on their ankles and wrist continuously for 1–5 days to measure LA from movements during sleep. Regression models with built-in feature selection were used to determine the most relevant LA features and the association to each measure of impairment. Results LA features were related to measures...

Sensors, 2020

In recent years, human activity recognition has become a hot topic inside the scientific community. The reason to be under the spotlight is its direct application in multiple domains, like healthcare or fitness. Additionally, the current worldwide use of smartphones makes it particularly easy to get this kind of data from people in a non-intrusive and cheaper way, without the need for other wearables. In this paper, we introduce our orientation-independent, placement-independent and subject-independent human activity recognition dataset. The information in this dataset is the measurements from the accelerometer, gyroscope, magnetometer, and GPS of the smartphone. Additionally, each measure is associated with one of the four possible registered activities: inactive, active, walking and driving. This work also proposes asupport vector machine (SVM) model to perform some preliminary experiments on the dataset. Considering that this dataset was taken from smartphones in their actual use...

PLOS digital health, 2022

Stroke rehabilitation seeks to accelerate motor recovery by training functional activities, but may have minimal impact because of insufficient training doses. In animals, training hundreds of functional motions in the first weeks after stroke can substantially boost upper extremity recovery. The optimal quantity of functional motions to boost recovery in humans is currently unknown, however, because no practical tools exist to measure them during rehabilitation training. Here, we present PrimSeq, a pipeline to classify and count functional motions trained in stroke rehabilitation. Our approach integrates wearable sensors to capture upper-body motion, a deep learning model to predict motion sequences, and an algorithm to tally motions. The trained model accurately decomposes rehabilitation activities into elemental functional motions, outperforming competitive machine learning methods. Prim-Seq furthermore quantifies these motions at a fraction of the time and labor costs of human experts. We demonstrate the capabilities of PrimSeq in previously unseen stroke patients with a range of upper extremity motor impairment. We expect that our methodological advances will support the rigorous measurement required for quantitative dosing trials in stroke rehabilitation.

Sensors, 2019

Recently, egocentric activity recognition has attracted considerable attention in the pattern recognition and artificial intelligence communities because of its wide applicability in medical care, smart homes, and security monitoring. In this study, we developed and implemented a deep-learning-based hierarchical fusion framework for the recognition of egocentric activities of daily living (ADLs) in a wearable hybrid sensor system comprising motion sensors and cameras. Long short-term memory (LSTM) and a convolutional neural network are used to perform egocentric ADL recognition based on motion sensor data and photo streaming in different layers, respectively. The motion sensor data are used solely for activity classification according to motion state, while the photo stream is used for further specific activity recognition in the motion state groups. Thus, both motion sensor data and photo stream work in their most suitable classification mode to significantly reduce the negative in...

Arabian journal for science and engineering, 2021

This paper provides a comparative study on the different techniques of classifying human activities that are performed using body-worn miniature inertial and magnetic sensors. The classification techniques implemented and compared in this study are: Bayesian decision making (BDM), a rule-based algorithm (RBA) or decision tree, the least-squares method (LSM), the k-nearest neighbor algorithm (k-NN), dynamic time warping (DTW), support vector machines (SVM), and artificial neural networks (ANN). Human activities are classified using five sensor units worn on the chest, the arms, and the legs. Each sensor unit comprises a tri-axial gyroscope, a tri-axial accelerometer, and a tri-axial magnetometer. A feature set extracted from the raw sensor data using principal component analysis (PCA) is used in the classification process. A performance comparison of the classification techniques is provided in terms of their correct differentiation rates, confusion matrices, and computational cost, as well as their preprocessing, training, and storage requirements. Three different cross-validation techniques are employed to validate the classifiers. The results indicate that in general, BDM results in the highest correct classification rate with relatively small computational cost.

Sensors, 2023

This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY

ACM Transactions in Embedded Computing Systems, 2022

Human Activity Recognition (HAR) based on inertial data is an increasingly diffused task on embedded devices, from smartphones to ultra low-power sensors. Due to the high computational complexity of deep learning models, most embedded HAR systems are based on simple and not-so-accurate classic machine learning algorithms. This work bridges the gap between on-device HAR and deep learning, proposing a set of efficient 1-dimensional Convolutional Neural Networks (1D CNNs) that can be deployed on general purpose microcontrollers (MCUs). Our CNNs are obtained combining hyper-parameters optimization with sub-byte and mixed-precision quantization, in order to find good trade-offs between classification results and memory occupation. Moreover, we also leverage adaptive inference as an orthogonal optimization to tune the inference complexity at runtime based on the processed input, hence producing a more flexible HAR system. With experiments on 4 datasets, and targeting an ultra-low-power RISC-V MCU, we show that: i) We are able to obtain a rich set of Pareto-optimal CNNs for HAR, spanning more than 1 order of magnitude in terms of memory, latency, and energy consumption; ii) Thanks to adaptive inference, we can derive > 20 runtime operating modes starting from a single CNN, differing by up to 10% in classification scores and by more than 3x in inference complexity, with a limited memory overhead; iii) On 3 of the 4 benchmarks, we outperform all previous deep learning methods, while reducing the memory occupation by more than 100x. The few methods that obtain better performance (both shallow or deep) are not compatible with MCU deployment. iv) All our CNNs are compatible with real-time on-device HAR, achieving an inference latency that ranges between 9 s and 16ms. Their memory occupation varies in 0.05-23.17 kB, and their energy consumption in 0.05 J and 61.59 J, allowing years of continuous operation on a small battery supply.

Journal of Neuromuscular Diseases, 2021

In 2019, stride velocity 95th centile (SV95C) became the first wearable-derived digital clinical outcome assessment (COA) qualified by the European Medicines Agency (EMA) for use as a secondary endpoint in trials for Duchenne muscular dystrophy. SV95C was approved via the EMA’s qualification pathway for novel methodologies for medicine development, which is a voluntary procedure for assessing the regulatory acceptability of innovative methods used in pharmaceutical research and development. SV95C is an objective, real-world digital ambulation measure of peak performance, representing the speed of the fastest strides taken by the wearer over a recording period of 180 hours. SV95C is correlated with traditional clinic-based assessments of motor function and has greater sensitivity to clinical change over 6 months than other wearable-derived stride variables, for example, median stride length or velocity. SV95C overcomes many limitations of episodic, clinic-based motor function testing...