Inertial Body Sensor Research Articles

Magnetic-Free Quaternion-Based Robust Unscented Kalman Filter for Upper Limb Kinematic Analysis

Human motion analysis is a field of increasing interest. In most applications, an accurate estimation of joint angles is needed. Inertial measurement units (IMUs) represent a very promising technology for motion capture. Unfortunately, their performances are strongly influenced by different sources of error (e.g., gyroscopes biases). To limit the effect of these sources of error, several algorithms (usually defined as “sensor fusion” algorithms) are used. The proposed work aims to validate a magnetic-free quaternion-based robust unscented Kalman filter (UKF) for upper limb kinematic analysis (i.e., estimation of the articular joint angles), using an inertial body sensor network, in a set of complex 3-D movements (i.e., the yoga poses). The algorithm was tested on ten expert yoga practitioners during the execution of the sun salutation sequence. Joint angle estimations were compared with the ones obtained from an optoelectronic reference system by evaluating the root mean square errors (RMSEs) and Pearson’s correlation coefficients. The achieved worst case was 8.41°, while the best one was 2.96° for RMSEs mean values. The accuracy of the algorithm was further confirmed by the high values of Pearson’s correlation coefficients (lowest mean value of 0.80). In addition, the comparison with one of the most used filters for attitude estimation [extended Kalman filter (EKF)] showed no significant differences between the two methods. Up to authors knowledge, the scientific literature lacks the development of a complete magnetic-free version of the UKF for human motion capture and analysis in complex movements, such as the one introduced here. The proposed work aims at filling this gap.

HARGRURNN: Human activity recognition using inertial body sensor gated recurrent units recurrent neural network

Internet of Things (IoT), wearable devices, and smartphones are more ubiquitous and available at a reasonable cost. Fitbit, Microsoft Band, Apple Watch, MI band, and Smartphone apps namely Strava and Runkeeper are some of the commercial products which are already available in the market. These products are embedded with the sensors that facilitate them to regularly sense and capture the environmental, physiological, functional data for applications in healthcare, wellbeing, and sports. These sensing devices have mediocre computing capabilities for data processing and transfer. The size of devices is so compact that can be worn on the body put in the pocket or locate in the house. This large scale recorded a wealth of physiological data require an efficient and meaningful interpretation as well as the proper and proficient method of analysis and the classification of data.Human Activity Recognition is classified into two features say Shallow and Deep features. Shallow features are extracted conventionally with the help of a simple machine learning approach. The large-scale time series situ data which require high computation power, processing capabilities, activities, and real-time classification, Deep Learning (DL) is the promising technique to deal with these activities.Weexercise the triaxial Accelerometers and triaxial Gyroscope to capture the data for HAR. The data set measured by the inertial sensor is then divided into the segments of 4 to 10s. The performance will be shown on the parameters of precision (%), Recall (%) and Accuracy (%).Parameter optimization and extraction of the segments are the two areas where we can carry forward our research.

Robust Human Activity Recognition Using Multimodal Feature-Level Fusion

Automated recognition of human activities or actions has great significance as it incorporates wide-ranging applications, including surveillance, robotics, and personal health monitoring. Over the past few years, many computer vision-based methods have been developed for recognizing human actions from RGB and depth camera videos. These methods include space-time trajectory, motion encoding, key poses extraction, space-time occupancy patterns, depth motion maps, and skeleton joints. However, these camera-based approaches are affected by background clutter and illumination changes and applicable to a limited field of view only. Wearable inertial sensors provide a viable solution to these challenges but are subject to several limitations such as location and orientation sensitivity. Due to the complementary trait of the data obtained from the camera and inertial sensors, the utilization of multiple sensing modalities for accurate recognition of human actions is gradually increasing. This paper presents a viable multimodal feature-level fusion approach for robust human action recognition, which utilizes data from multiple sensors, including RGB camera, depth sensor, and wearable inertial sensors. We extracted the computationally efficient features from the data obtained from RGB-D video camera and inertial body sensors. These features include densely extracted histogram of oriented gradient (HOG) features from RGB/depth videos and statistical signal attributes from wearable sensors data. The proposed human action recognition (HAR) framework is tested on a publicly available multimodal human action dataset UTD-MHAD consisting of 27 different human actions. K-nearest neighbor and support vector machine classifiers are used for training and testing the proposed fusion model for HAR. The experimental results indicate that the proposed scheme achieves better recognition results as compared to the state of the art. The feature-level fusion of RGB and inertial sensors provides the overall best performance for the proposed system, with an accuracy rate of 97.6%.

Toward Pervasive Gait Analysis With Wearable Sensors: A Systematic Review.

After decades of evolution, measuring instruments for quantitative gait analysis have become an important clinical tool for assessing pathologies manifested by gait abnormalities. However, such instruments tend to be expensive and require expert operation and maintenance besides their high cost, thus limiting them to only a small number of specialized centers. Consequently, gait analysis in most clinics today still relies on observation-based assessment. Recent advances in wearable sensors, especially inertial body sensors, have opened up a promising future for gait analysis. Not only can these sensors be more easily adopted in clinical diagnosis and treatment procedures than their current counterparts, but they can also monitor gait continuously outside clinics - hence providing seamless patient analysis from clinics to free-living environments. The purpose of this paper is to provide a systematic review of current techniques for quantitative gait analysis and to propose key metrics for evaluating both existing and emerging methods for qualifying the gait features extracted from wearable sensors. It aims to highlight key advances in this rapidly evolving research field and outline potential future directions for both research and clinical applications.

Causality Analysis of Inertial Body Sensors for Multiple Sclerosis Diagnostic Enhancement.

Inertial body sensors have emerged in recent years as an effective tool for evaluating mobility impairment resulting from various diseases, disorders, and injuries. For example, body sensors have been used in 6-min walk (6MW) tests for multiple sclerosis (MS) patients to identify gait features useful in the study, diagnosis, and tracking of the disease. However, most studies to date have focused on features localized to the lower or upper extremities and do not provide a holistic assessment of mobility. This paper presents a causality analysis method focused on the coordination between extremities to identify subtle whole-body mobility impairment that may aid disease diagnosis. This method was developed for and utilized in an MS pilot study with 41 subjects (28 persons with MS (PwMS) and 13 healthy controls) performing 6MW tests. Compared with existing methods, the causality analysis provided better discrimination between healthy controls and PwMS and a deeper understanding of MS disease impact on mobility.

Piecewise Linear Dynamical Model for Action Clustering from Real-World Deployments of Inertial Body Sensors

Human motion has been reported as having great relevance to various disease, disorder, injuries and emotional state. Therefore, motion assessment using inertial body sensor networks (BSNs) is gaining popularity as an outcome measure in clinical study and neuroscience research. The efficacy of motion assessment heavily relies on the accurate temporal clustering of human motion into actions on various time scales. However, two human factors in real-world deployments of inertial BSNs make such motion assessment challenging: mounting errors (where sensor displacement and orientation do not match what is assumed by processing algorithms) and insecure mounting (where sensors are loosely worn causing them to shake during operations). In order to enhance the robustness of human actions clutsering from real-world BSN data, this work leverages dynamical systems modeling with the considerations of human factors. By proposing a computational body-model framework called the piecewise linear dynamical model (PLDM), we derive a robust method to segment time series data of inertial BSNs in real-world deployment with human factors into motion primitives and actions. We test the proposed method on three different inertial BSN datasets, extract actions on different temporal scales and recognize the actions into clusters. The experimental results demonstrate the effectiveness of our approach.

Quaternion-valued echo state networks.

Quaternion-valued echo state networks (QESNs) are introduced to cater for 3-D and 4-D processes, such as those observed in the context of renewable energy (3-D wind modeling) and human centered computing (3-D inertial body sensors). The introduction of QESNs is made possible by the recent emergence of quaternion nonlinear activation functions with local analytic properties, required by nonlinear gradient descent training algorithms. To make QENSs second-order optimal for the generality of quaternion signals (both circular and noncircular), we employ augmented quaternion statistics to introduce widely linear QESNs. To that end, the standard widely linear model is modified so as to suit the properties of dynamical reservoir, typically realized by recurrent neural networks. This allows for a full exploitation of second-order information in the data, contained both in the covariance and pseudocovariances, and a rigorous account of second-order noncircularity (improperness), and the corresponding power mismatch and coupling between the data components. Simulations in the prediction setting on both benchmark circular and noncircular signals and on noncircular real-world 3-D body motion data support the analysis.

Improving Human Action Recognition Using Fusion of Depth Camera and Inertial Sensors

This paper presents a fusion approach for improving human action recognition based on two differing modality sensors consisting of a depth camera and an inertial body sensor. Computationally efficient action features are extracted from depth images provided by the depth camera and from accelerometer signals provided by the inertial body sensor. These features consist of depth motion maps and statistical signal attributes. For action recognition, both feature-level fusion and decision-level fusion are examined by using a collaborative representation classifier. In the feature-level fusion, features generated from the two differing modality sensors are merged before classification, while in the decision-level fusion, the Dempster–Shafer theory is used to combine the classification outcomes from two classifiers, each corresponding to one sensor. The introduced fusion framework is evaluated using the Berkeley multimodal human action database. The results indicate that because of the complementary aspect of the data from these sensors, the introduced fusion approaches lead to 2% to 23% recognition rate improvements depending on the action over the situations when each sensor is used individually.

Correlations between Inertial Body Sensor Measures and Clinical Measures in Multiple Sclerosis

Gait assessment using inertial body sensors is becoming popular as an outcome measure in multiple sclerosis (MS) research, supplementing clinical observations and patient-reported outcomes with precise, objective measures. Although numerous research reports have demonstrated the performance of inertial measures in distinguishing healthy controls and MS subjects, the relationship between these measures and the impact of MS on gait impairment remains poorly understood. In contrast, although clinical evaluation has limited variability in scores, it is meaningful and interpretable for clinicians.Therefore, this paper investigates correlations between two inertial measures and three clinical measures of walking ability. The clinical measures are the MS Walking Scale (MSWS-12), the Expanded Disability Status Scale (EDSS), and the six minute walk (6MW) distance. The inertial measures are the double stance time to single stance time ratio (DST/SST) and the causality index, both of which have been proven effective in MS gait assessment in previous work.28 MS subjects and 13 healthy controls were recruited from an MS outpatient clinic. Most correlations among measures were strong and significant. Experimental results suggested that combining all five measures may improve separability performance for tracking MS disease progression.

Quaternion Reproducing Kernel Hilbert Spaces: Existence and Uniqueness Conditions

The existence and uniqueness conditions of quaternion reproducing kernel Hilbert spaces (QRKHS) are established in order to provide a mathematical foundation for the development of quaternion-valued kernel learning algorithms. This is achieved through a rigorous account of left quaternion Hilbert spaces, which makes it possible to generalise standard RKHS to quaternion RKHS. Quaternion versions of the Riesz representation and Moore-Aronszajn theorems are next introduced, thus underpinning kernel estimation algorithms operating on quaternion-valued feature spaces. The difference between the proposed quaternion kernel concept and the existing real and vector approaches is also established in terms of both theoretical advantages and computational complexity. The enhanced estimation ability of the so-introduced quaternion-valued kernels over their real- and vector-valued counterparts is validated through kernel ridge regression applications. Simulations on real world 3D inertial body sensor data and nonlinear channel equalisation using novel quaternion cubic and Gaussian kernels support the approach.

Laboratory Validation of Inertial Body Sensors to Detect Cigarette Smoking Arm Movements

Cigarette smoking remains the leading cause of preventable death in the United States. Traditional in-clinic cessation interventions may fail to intervene and interrupt the rapid progression to relapse that typically occurs following a quit attempt. The ability to detect actual smoking behavior in real-time is a measurement challenge for health behavior research and intervention. The successful detection of real-time smoking through mobile health (mHealth) methodology has substantial implications for developing highly efficacious treatment interventions. The current study was aimed at further developing and testing the ability of inertial sensors to detect cigarette smoking arm movements among smokers. The current study involved four smokers who smoked six cigarettes each in a laboratory-based assessment. Participants were outfitted with four inertial body movement sensors on the arms, which were used to detect smoking events at two levels: the puff level and the cigarette level. Two different algorithms (Support Vector Machines (SVM) and Edge-Detection based learning) were trained to detect the features of arm movement sequences transmitted by the sensors that corresponded with each level. The results showed that performance of the SVM algorithm at the cigarette level exceeded detection at the individual puff level, with low rates of false positive puff detection. The current study is the second in a line of programmatic research demonstrating the proof-of-concept for sensor-based tracking of smoking, based on movements of the arm and wrist. This study demonstrates efficacy in a real-world clinical inpatient setting and is the first to provide a detection rate against direct observation, enabling calculation of true and false positive rates. The study results indicate that the approach performs very well with some participants, whereas some challenges remain with participants who generate more frequent non-smoking movements near the face. Future work may allow for tracking smoking in real-world environments, which would facilitate developing more effective, just-in-time smoking cessation interventions.

Multikernel least mean square algorithm.

The multikernel least-mean-square algorithm is introduced for adaptive estimation of vector-valued nonlinear and nonstationary signals. This is achieved by mapping the multivariate input data to a Hilbert space of time-varying vector-valued functions, whose inner products (kernels) are combined in an online fashion. The proposed algorithm is equipped with novel adaptive sparsification criteria ensuring a finite dictionary, and is computationally efficient and suitable for nonstationary environments. We also show the ability of the proposed vector-valued reproducing kernel Hilbert space to serve as a feature space for the class of multikernel least-squares algorithms. The benefits of adaptive multikernel (MK) estimation algorithms are illuminated in the nonlinear multivariate adaptive prediction setting. Simulations on nonlinear inertial body sensor signals and nonstationary real-world wind signals of low, medium, and high dynamic regimes support the approach.

Balance and knee extensibility evaluation of hemiplegic gait using an inertial body sensor network

BackgroundMost hemiplegic patients have difficulties in their balance and posture control while walking because of the asymmetrical posture and the abnormal body balance. The assessment of rehabilitation of hemiplegic gait is usually made by doctors using clinical scale, but it is difficult and could not be used frequently. It is therefore needed to quantitatively analyze the characteristics of hemiplegic gait. Thus the assessment would be simple, and real-time evaluation of rehabilitation could be carried out.MethodsTwenty subjects (ten hemiplegic patients, ten normal subjects) were recruited. The subjects walked straight for five meters at their self-selected comfortable speed towards a target line on the floor.Xsens MTx motion trackers were used for acquiring gestures of body segments to estimate knee joint angles and identify gait cycles. A practical method for data acquisition that does not need to obtain accurate distances between a knee joint and its corresponding sensors is presented.ResultsThe results showed that there were significant differences between the two groups in the three nominated angle amplitudes. The mean values of balance level of each parameter in hemiplegic gait and normal gait were: 0.21 versus 0.01, 0.18 versus 0.03, and 0.92 versus 0.03, respectively. The mean values of added angles of each parameter in hemiplegic gait and normal gait were: 74.64 versus 91.31, -76.48 versus −132.4, and 6.77 versus 35.74.ConclusionsIt was concluded that the wearable bio-motion acquisition platform provided a practical approach that was effective in discriminating gait symptoms between hemiplegic and asymptomatic subjects. The extensibility of hemiplegic patients’ lower limbs was significantly lower than that of normal subjects, and the hemiplegic gait had worse balance level compared with normal gait. The effect of rehabilitation training of hemiplegic gait could be quantitatively analyzed.

Characterising and minimising sources of error in inertial body sensor networks

Body sensor network BSN applications depend on accurate and precise data from body-worn devices, but issues related to sensor variations, body mounting variations, integration drift and node-to-node synchronisation can dramatically impact the quality and reliability of collected data and, ultimately, application fidelity. Characterising and addressing these sources of error which are both static and dynamic e.g. sensors suffer from static manufacturing variability and dynamic environmental impacts - within the context of application requirements is therefore necessary for the viability of such applications. This work characterises and addresses errors related to node synchronisation, sensor and mounting calibration and integration drift on a case study application - knee joint angle as measured during walking by the TEMPO 3.1 inertial BSN platform. Using the Vicon® optical motion capture system to provide ground truth, synchronisation, calibration and drift error are quantied, and the efficacy of solutions for reducing such errors is evaluated.

Exploration and Implementation of a Pre-Impact Fall Recognition Method Based on an Inertial Body Sensor Network

The unintentional injuries due to falls in elderly people give rise to a multitude of health and economic problems due to the growing aging population. The use of early pre-impact fall alarm and self-protective control could greatly reduce fall injuries. This paper aimed to explore and implement a pre-impact fall recognition/alarm method for free-direction fall activities based on understanding of the pre-impact lead time of falls and the angle of body postural stability using an inertial body sensor network. Eight healthy Asian adult subjects were arranged to perform three kinds of daily living activities and three kinds of fall activities. Nine MTx sensor modules were used to measure the body segmental kinematic characteristics of each subject for pre-impact fall recognition/alarm. Our analysis of the kinematic features of human body segments showed that the chest was the optimal sensor placement for an early pre-impact recognition/alarm (i.e., prediction/alarm of a fall event before it happens) and post-fall detection (i.e., detection of a fall event after it already happened). Furthermore, by comparative analysis of threshold levels for acceleration and angular rate, two acceleration thresholds were determined for early pre-impact alarm (7 m/s/s) and post-fall detection (20 m/s/s) under experimental conditions. The critical angles of postural stability of torso segment in three kinds of fall activities (forward, sideway and backward fall) were determined as 23.9 ± 3.3, 49.9 ± 4.1 and 9.9 ± 2.5 degrees, respectively, and the relative average pre-impact lead times were 329 ± 21, 265 ± 35 and 257 ± 36 ms. The results implied that among the three fall activities the sideway fall was associated with the largest postural stability angle and the forward fall was associated with the longest time to adjust body angle to avoid the fall; the backward fall was the most difficult to avoid among the three kinds of fall events due to the toughest combination of shortest lead time and smallest angle of postural stability which made it difficult for the self-protective control mechanism to adjust the body in time to avoid falling down.

Analyzing human gait and posture by combining feature selection and kernel methods

This paper evaluates a set of computational algorithms for the automatic estimation of human postures and gait properties from signals provided by an inertial body sensor. The use of a single sensor device imposes limitations for the automatic estimation of relevant properties, like step length and gait velocity, as well as for the detection of standard postures like sitting or standing. Moreover, the exact location and orientation of the sensor are also a common restriction that is relaxed in this study. Based on accelerations provided by a sensor, known as the ‘9×2’, three approaches are presented extracting kinematic information from the user motion and posture. First, a two-phases procedure implementing feature extraction and support vector machine based classification for daily living activity monitoring is presented. Second, support vector regression is applied on heuristically extracted features for the automatic computation of spatiotemporal properties during gait. Finally, sensor information is interpreted as an observation of a particular trajectory of the human gait dynamical system, from which a reconstruction space is obtained, and then transformed using standard principal components analysis, finally support vector regression is used for prediction. Daily living activities are detected and spatiotemporal parameters of human gait are estimated using methods sharing a common structure based on feature extraction and kernel methods. The approaches presented are susceptible to be used for medical purposes.