Single Tri-axial Accelerometer Research Articles

Detecting Steps Walking at very Low Speeds Combining Outlier Detection, Transition Matrices and Autoencoders from Acceleration Patterns.

In this paper, we develop and validate a new algorithm to detect steps while walking at speeds between 30 and 40 steps per minute based on the data sensed from a single tri-axial accelerometer. The algorithm concatenates three consecutive phases. First, an outlier detection is performed on the sensed data based on the Mahalanobis distance to pre-detect candidate points in the acceleration time series that may contain a ground contact segment of data while walking. Second, the acceleration segment around the pre-detected point is used to calculate the transition matrix in order to capture the time dependencies. Finally, autoencoders, trained with data segments containing ground contact transition matrices from acceleration series from labeled steps are used to reconstruct the computed transition matrices at each pre-detected point. A similarity index is used to assess if the pre-selected point contains a true step in the 30–40 steps per minute speed range. Our experimental results, based on a database from three different participants performing similar activities to the target one, are able to achieve a recall = 0.88 with precision = 0.50 improving the results when directly applying the autoencoders to acceleration patterns (recall = 0.77 with precision = 0.50).

Developing Fine-Grained Actigraphies for Rheumatoid Arthritis Patients from a Single Accelerometer Using Machine Learning.

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.

Better than counting seconds: Identifying fallers among healthy elderly using fusion of accelerometer features and dual-task Timed Up and Go.

Devices and sensors for identification of fallers can be used to implement actions to prevent falls and to allow the elderly to live an independent life while reducing the long-term care costs. In this study we aimed to investigate the accuracy of Timed Up and Go test, for fallers’ identification, using fusion of features extracted from accelerometer data. Single and dual tasks TUG (manual and cognitive) were performed by a final sample (94% power) of 36 community dwelling healthy older persons (18 fallers paired with 18 non-fallers) while they wear a single triaxial accelerometer at waist with sampling rate of 200Hz. The segmentation of the TUG different trials and its comparative analysis allows to better discriminate fallers from non-fallers, while conventional functional tests fail to do so. In addition, we show that the fusion of features improve the discrimination power, achieving AUC of 0.84 (Sensitivity = Specificity = 0.83, 95% CI 0.62-0.91), and demonstrating the clinical relevance of the study. We concluded that features extracted from segmented TUG trials acquired with dual tasks has potential to improve performance when identifying fallers via accelerometer sensors, which can improve TUG accuracy for clinical and epidemiological applications.

Home detection of freezing of gait using support vector machines through a single waist-worn triaxial accelerometer.

Among Parkinson’s disease (PD) symptoms, freezing of gait (FoG) is one of the most debilitating. To assess FoG, current clinical practice mostly employs repeated evaluations over weeks and months based on questionnaires, which may not accurately map the severity of this symptom. The use of a non-invasive system to monitor the activities of daily living (ADL) and the PD symptoms experienced by patients throughout the day could provide a more accurate and objective evaluation of FoG in order to better understand the evolution of the disease and allow for a more informed decision-making process in making adjustments to the patient’s treatment plan. This paper presents a new algorithm to detect FoG with a machine learning approach based on Support Vector Machines (SVM) and a single tri-axial accelerometer worn at the waist. The method is evaluated through the acceleration signals in an outpatient setting gathered from 21 PD patients at their home and evaluated under two different conditions: first, a generic model is tested by using a leave-one-out approach and, second, a personalised model that also uses part of the dataset from each patient. Results show a significant improvement in the accuracy of the personalised model compared to the generic model, showing enhancement in the specificity and sensitivity geometric mean (GM) of 7.2%. Furthermore, the SVM approach adopted has been compared to the most comprehensive FoG detection method currently in use (referred to as MBFA in this paper). Results of our novel generic method provide an enhancement of 11.2% in the GM compared to the MBFA generic model and, in the case of the personalised model, a 10% of improvement with respect to the MBFA personalised model. Thus, our results show that a machine learning approach can be used to monitor FoG during the daily life of PD patients and, furthermore, personalised models for FoG detection can be used to improve monitoring accuracy.

Adaptive sliding window segmentation for physical activity recognition using a single tri-axial accelerometer

Previous studies on physical activity recognition have utilized various fixed window sizes for signal segmentation targeting specific activities. Naturally, an optimum window size varies depending on the characteristics of activity signals and fixed window size will not produce good segmentation for all activities. This paper presents a novel approach to activity signal segmentation for physical activity recognition. Central to the approach is that the window size is adaptively adjusted according to the probability of the signal belongs to a particular activity to achieve the most effective segmentation. In addition, an activity transition diagram for activity recognition is developed to validate the activity transition and improve recognition accuracy. The adaptive sliding window segmentation algorithm and the role of activity transition diagram are described in the context of physical activity recognition. The approach recognizes not only well defined static and dynamic activities, but also transitional activities. The presented approach has been implemented, evaluated and compared with an existing state-of-the-art approach by using internal and public datasets which contains activity signals of dynamic, static and transitional activities. Results have shown that the proposed adaptive sliding window segmentation achieves overall accuracy of 95.4% in all activities considered in the experiments compared to the existing approach which achieved an overall accuracy of 89.9%. The proposed approach achieved an overall accuracy of 96.5% compared to 91.9% overall accuracy with the existing approach when tested on the public dataset.

Human Movement Recognition Based on the Stochastic Characterisation of Acceleration Data

Human activity recognition algorithms based on information obtained from wearable sensors are successfully applied in detecting many basic activities. Identified activities with time-stationary features are characterised inside a predefined temporal window by using different machine learning algorithms on extracted features from the measured data. Better accuracy, precision and recall levels could be achieved by combining the information from different sensors. However, detecting short and sporadic human movements, gestures and actions is still a challenging task. In this paper, a novel algorithm to detect human basic movements from wearable measured data is proposed and evaluated. The proposed algorithm is designed to minimise computational requirements while achieving acceptable accuracy levels based on characterising some particular points in the temporal series obtained from a single sensor. The underlying idea is that this algorithm would be implemented in the sensor device in order to pre-process the sensed data stream before sending the information to a central point combining the information from different sensors to improve accuracy levels. Intra- and inter-person validation is used for two particular cases: single step detection and fall detection and classification using a single tri-axial accelerometer. Relevant results for the above cases and pertinent conclusions are also presented.

머신러닝을 이용한 앉은 자세 분류 연구

According to recent studies, poor sitting posture of the spine has been shown to lead to a variety of spinal disorders. For this reason, it is important to measure the sitting posture. We proposed a strategy for classification of sitting posture using machine learning. We retrieved acceleration data from single tri-axial accelerometer attached on the back of the subject’s neck in 5-types of sitting posture. 6 subjects without any spinal disorder were participated in this experiment. Acceleration data were transformed to the feature vectors of principle component analysis. Support vector machine (SVM) and K-means clustering were used to classify sitting posture with the transformed feature vectors. To evaluate performance, we calculated the correct rate for each classification strategy. Although the correct rate of SVM in sitting back arch was lower than that of K-means clustering by 2.0%, SVM’s correct rate was higher by 1.3%, 5.2%, 16.6%, 7.1% in a normal posture, sitting front arch, sitting cross-legged, sitting leaning right, respectively. In conclusion, the overall correction rates were 94.5% and 88.84% in SVM and K-means clustering respectively, which means that SVM have more advantage than K-means method for classification of sitting posture.

Test-Retest Reliability and Concurrent Validity of a Single Tri-Axial Accelerometer-Based Gait Analysis in Older Adults with Normal Cognition.

ObjectiveWe investigated the concurrent validity and test-retest reliability of spatio-temporal gait parameters measured with a single tri-axial accelerometer (TAA), determined the optimal number of steps required for obtaining acceptable levels of reliability, and compared the validity and reliability of the estimated gait parameters across the three reference axes of the TAA.MethodsA total of 82 cognitively normal elderly participants walked around a 40-m long round walkway twice wearing a TAA at their center of body mass. Gait parameters such as cadence, gait velocity, step time, step length, step time variability, and step time asymmetry were estimated from the low pass-filtered signal of the TAA. The test-retest reliability and concurrent validity with the GAITRite® system were evaluated for the estimated gait parameters.ResultsGait parameters using signals from the vertical axis showed excellent reliability for all gait parameters; the intraclass correlation coefficient (ICC) was 0.79–0.90. A minimum of 26 steps and 14 steps were needed to achieve excellent reliability in step time variability and step time asymmetry, respectively. A strong level of agreement was seen for the basic gait parameters between the TAA and GAITRiteⓇ (ICC = 0.91–0.96).ConclusionsThe measurement of gait parameters of elderly individuals with normal cognition using a TAA placed on the body’s center of mass was reliable and showed superiority over the GAITRiteⓇ with regard to gait variability and asymmetry. The TAA system was a valid tool for measuring basic gait parameters. Considering its wearability and low price, the TAA system may be a promising alternative to the pressure sensor walkway system for measuring gait parameters.

ASSESSMENT OF POSTURAL INSTABILITY IN PATIENTS WITH A NEUROLOGICAL DISORDER USING A TRI-AXIAL ACCELEROMETER

<span lang="EN-US">Current techniques for quantifying human postural stability during quiet standing have several limitations. The main problem is that only two movement variables are evaluated, though a better description of complex three-dimensional (3-D) movements can be provided with the use of three variables. A single tri-axial accelerometer placed on the trunk was used to measure 3-D data.<br />We are able to evaluate 3-D movements using a method based on the volume of confidence ellipsoid (VE) of the set of points obtained by plotting three accelerations against each other. Our method was used to identify and evaluate pathological balance control. In this study, measurements were made of patients with progressive cerebellar ataxia, and also control measurements of healthy subjects, and a statistical analysis was performed. The results show that the VEs of the neurological disorder patients are significantly larger than the VEs of the healthy subjects. It can be seen that the quantitative method based on VE is very sensitive for identifying changes in stability, and that it is able to distinguish between neurological disorder patients and healthy subjects.<br /></span>

Human activity recognition using quasiperiodic time series collected from a single tri-axial accelerometer

The current generation of portable mobile devices incorporates various types of sensors that open up new areas for the analysis of human behavior. In this paper, we propose a method for human physical activity recognition using time series, collected from a single tri-axial accelerometer of a smartphone. Primarily, the method solves a problem of online time series segmentation, assuming that each meaningful segment corresponds to one fundamental period of motion. To extract the fundamental period we construct the phase trajectory matrix, applying the technique of principal component analysis. The obtained segments refer to various types of human physical activity. To recognize these activities we use the k-nearest neighbor algorithm and neural network as an alternative. We verify the accuracy of the proposed algorithms by testing them on the WISDM dataset of labeled accelerometer time series from thirteen users. The results show that our method achieves high precision, ensuring nearly 96 % recognition accuracy when using the bunch of segmentation and k-nearest neighbor algorithms.

Human Activity Recognition Based on the Hierarchical Feature Selection and Classification Framework

Human activity recognition via triaxial accelerometers can provide valuable information for evaluating functional abilities. In this paper, we present an accelerometer sensor-based approach for human activity recognition. Our proposed recognition method used a hierarchical scheme, where the recognition of ten activity classes was divided into five distinct classification problems. Every classifier used the Least Squares Support Vector Machine (LS-SVM) and Naive Bayes (NB) algorithm to distinguish different activity classes. The activity class was recognized based on the mean, variance, entropy of magnitude, and angle of triaxial accelerometer signal features. Our proposed activity recognition method recognized ten activities with an average accuracy of 95.6% using only a single triaxial accelerometer.

Validity of a trunk-mounted accelerometer to assess peak accelerations during walking, jogging and running

The purpose of this study was to validate peak acceleration data from an accelerometer contained within a wearable tracking device while walking, jogging and running. Thirty-nine participants walked, jogged and ran on a treadmill while 10 peak accelerations per movement were obtained (n = 390). A single triaxial accelerometer measured resultant acceleration during all movements. To provide a criterion measure of acceleration, a 12-camera motion analysis (MA) system tracked the position of a retro-reflective marker affixed to the wearable tracking device. Peak raw acceleration recorded by the accelerometer significantly overestimated peak MA acceleration (P < 0.01). Filtering accelerometer data improved the relationship with the MA system (P < 0.01). However, only the 10 Hz and 8 Hz cut-off frequencies significantly reduced the errors found. The walk movement demonstrated the highest accuracy, agreement and precision and the lowest relative errors. Linear increases in error were observed for jog compared with walk and for run compared to both other movements. As the magnitude of acceleration increased, the strength of the relationship between the accelerometer and the criterion measure decreased. These results indicate that filtered accelerometer data provide an acceptable means of assessing peak accelerations, in particular for walking and jogging.

Real-Time Physical Activity Recognition Using a Single Triaxial Accelerometer Based on HMM

This study focuses on physical activity recognition of elder people using a single triaxial accelerometer attached on waist. With accelerations data acquired continuously by a wearable wireless device, a motion pattern recognition method based on acceleration time series analysis of human activity states is proposed. We use Hidden Markov Model (HMM) to identify physical activity series. Experimental result shows that the activity recognition rate is 95.7% using this method, which can play an important role in the adjuvant therapy.

Within trial validation and reliability of a single tri-axial accelerometer for gait assessment.

Gait is a sensitive biomarker of decline in both cognitive and physical function. Therefore, the collection of gait data is an important feature of clinical assessments. Accelerometer-based body worn sensors are quickly becoming the preferred tool for assessing gait because they are small, useable in a wide variety of settings, offer more continuous spatio-temporal analysis and are inexpensive when compared with traditional gait assessment methodologies. The purpose of this study was to determine the validity and within test reliability of a low cost body worn movement sensor with associated algorithms to assess gait in a large group of older and younger healthy adults. We collected gait data over intermittent walks on an instrumented walkway for a within trial validation and also used the same accelerometer derived gait data for a within test reliability analysis. ICCs for validation and reliability were >0.756 and >0.965, respectively.

Threshold selection and adjustment for online segmentation of one-stroke finger gestures using single tri-axial accelerometer

Online segmentation of continuous gestures is a key issue in accelerometer-based gesture recognition. The challenge is to segment short-duration, ambiguous, and individually different gestures by an accurate method for online applications. In this paper, we propose a threshold-based method for the online segmentation of one-stroke gestures using a single tri-axial accelerometer on a finger. The method first obtains a proper initial threshold based on Bayes decision theory to ensure accurate initial segmentation, and then an adaptation mechanism is designed to automatically adjust the threshold to improve user-dependent accuracy. A dataset of over 1,200 samples from 12 gestures and 25 subjects is created for evaluation. The results show that our method achieves high segmentation precision/recall and high user-dependent adaptability with low computational complexity.

Classification accuracy of a single tri-axial accelerometer for training background and experience level in runners

Accelerometers are increasingly used tools for gait analysis, but there remains a lack of research on their application to running and their ability to classify running patterns. The purpose of this study was to conduct an exploratory examination into the capability of a tri-axial accelerometer to classify runners of different training backgrounds and experience levels, according to their 3-dimensional (3D) accelerometer data patterns. Training background was examined with 14 competitive soccer players and 12 experienced marathon runners, and experience level was examined with 16 first-time and the same 12 experienced marathon runners. Discrete variables were extracted from 3D accelerations during a short run using root mean square, wavelet transformation, and autocorrelation procedures. A principal component analysis (PCA) was conducted on all variables, including gait speed to account for covariance. Eight PCs were retained, explaining 88% of the variance in the data. A stepwise discriminant analysis of PCs was used to determine the binary classification accuracy for training background and experience level, with and without the PC of Speed. With Speed, the accelerometer correctly classified 96% of runners for both training background and experience level. Without Speed, the accelerometer correctly classified 85% of runners based on training background, but only 68% based on experience level. These findings suggest that the accelerometer is effective in classifying athletes of different training backgrounds, but is less effective for classifying runners of different experience levels where gait speed is the primary discriminator.

Detecting Falls Using Accelerometers by Adaptive Thresholds in Mobile Devices

In this preliminary work, we presented an effective and efficient algorithm on adaptive thresholds to automatically recognize falls from acceleration signals collected by a single tri-axial accelerometer in a mobile phone. Initial thresholds depend mainly on their carrying position of mobile phones, and then are adjusted automatically by a self-learning process and a classification module. Our researches are designed for carrying phones in casual ways which has not been done in previous researches. An android-based software is designed for experiments and the results show the efficiency of our method and improvements have been made on detection accuracy after having the learning process.

The Validity of Gait Variability and Fractal Dynamics Obtained From a Single, Body-Fixed Triaxial Accelerometer

A single triaxial accelerometer has the ability to collect a large amount of continuous gait data to quantitatively assess the control of gait. Unfortunately, there is limited information on the validity of gait variability and fractal dynamics obtained from this device. The purpose of this study was to test the concurrent validity of the variability and fractal dynamic measures of gait provided by a triaxial accelerometer during a continuous 10 minute walk in older adults. Forty-one healthy older adults were fitted with a single triaxial accelerometer at the waist, as well as a criterion footswitch device before completing a ten minute overground walk. The concurrent validity of six outcome measures was examined using intraclass correlation coefficients (ICC) and 95% limits of agreement. All six dependent variables measured by the accelerometer displayed excellent agreement with the footswitch device. Mean parameters displayed the highest validity, followed by measures of variability and fractal dynamics in stride times and measures of variability and fractal dynamics in step times. These findings suggest that an accelerometer is a valid and unique device that has the potential to provide clinicians with valid quantitative data for assessing their clients' gait.

Uncertainty in Impact Identification Applied to a Commercial Wind Turbine Blade

This work evaluates the uncertainty of impact force and location estimates using an entropy-based impact identification algorithm applied to a commercial wind turbine blade. The effects of sensor placement, measurement directions and distance between impacts and sensor locations are studied. Results show that impacts to a 35m long wind turbine blade can be accurately located using a single tri-axial accelerometer regardless of sensor location. Uncertainties in impact force estimates are consistent across sensor locations. When omitting acceleration information in the spanwise direction, the bias and variance of force estimates is consistent, but when a single chan- nel of acceleration data is used, both increase somewhat. Impact force identification error was found to be uncorrelated with the distance between the impact and sensor location. The entropy of the estimated force time history, an indicator of the impulsivity of the estimate, was found to be a good indicator of the quality of force estimate. The bias and variance of impact force estimation error was found to be directly correlated with the entropy of the impact force estimate. When considering validation test data from all possible sensor configurations, the entropy of the recreated force estimates was a better indicator of the force magnitude prediction interval than was the specific sensor configuration. By classifying impact force estimates based upon entropy values, impact force prediction intervals were more precisely determined than when all validation impact data were considered at once.

Direct optimization of inference model for human activity and posture class recognition

The demand for ambulatory monitoring is rising due to the widespread adoption of home-based healthcare models for the increasing elderly population. Since monitoring patients makes it possible to provide remote medical services based on real-world contexts, context-aware technology is regarded as a crucial element in core module development for ubiquitous healthcare solutions. This paper describes the design and implementation of a class label tagging system for human activity recognition based on sensor data from a single tri-axial accelerometer attached to the waist of a human subject. Our human activity and posture classifier (APC) was designed to model more effectively an event that takes place over a period of time. Consequently, the APC problem becomes the process of tagging class labels onto sequential data from a single tri-axial accelerometer. This sequential tagging naturally led to our hypothesis that a linear combination of the classification model and the prior, approximated by observing a state transition event of human activity and posture (AP), may decrease the error rate of the system. Therefore, this paper aims at testing the aforementioned experimental hypothesis while explicating our method of direct optimization, also known as direct loss minimization. Our experimental hypothesis is based on the view that the task of labeling a series of class tags on a data object of the AP event should be supported by a statistical model capable of robustly capturing the probabilistic context influencing the generation of a set of features at a given discrete time. A radial basis function network algorithm was used for parameter estimation of a class likelihood inference model augmented by training a state transition model to capture the sequential nature of a posture or activity generation event. The linear combination requires our design step to have a further optimization, in addition to learning the parameters of each inference model, of the model parameters that represent the relative contribution to the decision process. Previous studies have elaborated on an effective approach to direct optimization of model parameters when an appropriate evaluation metric exists for a given problem domain. In our case, f-measure was used as an objective function of the direct optimization. We used local beam search to optimize directly the weight vector for the likelihood and the prior. Our method is tested on the set of nine-dimensional feature vectors contained in 69,376 lines of sequential AP data. The experiment shows that embedding prior probability into the recognition process decreases the error rate in a set of categories. Several experiments proved the efficiency of our proposed method and the usefulness of the system.