Accelerometer Signals Research Articles

Inertial sensor fusion for gait recognition with symmetric positive definite Gaussian kernels analysis

Wearable sensor-based gait recognition has received much interest because it is unobtrusive and is user friendly. Many research has been carried out in this area but conventional gait recognition methods are not free from drawbacks. In this paper, accelerometer and gyroscope signals representing gait movements are encoded using covariance matrices. The covariance matrices provide a compact and descriptive representation for the accelerometer and gyroscope signals. Non-singular covariance matrices are inherently Symmetric Positive Define (SPD) matrices. Interpreting such SPD matrices as points on the Riemannian manifold leads to increased performance. However, direct geodesic distance calculation for the matrix manifold may yield a suboptimal result. The proposed method solves this issue by embedding the manifold valued points to a higher dimensional Reproducing Kernel Hilbert Space (RKHS) via Positive Definite Gaussian Kernel functions. Extensive experiments have been conducted on three challenging benchmark datasets and a self-collected dataset. Experiment results testify the performance of the proposed RKHS embedding approach.

Deep Neural Network-Based Indoor Emergency Awareness Using Contextual Information From Sound, Human Activity, and Indoor Position on Mobile Device

Context-aware computing, which gathers information about the environment at a given time and adapts its behavior accordingly, can be effectively applied to a monitoring system that automatically detects and copes with emergency situations. In this paper, we propose an indoor emergency awareness alarm system using a deep neural network on a mobile device for at-risk people, such as the elderly and children. The proposed system detects emergency situations and quickly delivers the state of the person being monitored to a guardian using a mobile transmission system. To do this, three types of contextual information, including sound, human activity, and indoor location, are instantly recognized by the protected person’s mobile device. Both sound and human activity are used to improve the recognition accuracy of emergency situations provided to guardians’ mobile devices. Sound events are detected by a residual neural network, and human activities are recognized by applying the accelerometer and gyroscope signals of the mobile device to a deep spiking neural network. In addition, the indoor location where the emergency occurred is detected by the deep spiking neural network. The experimental results show the high accuracy of the indoor emergency awareness framework.

Glottal Aerodynamics Estimated From Neck-Surface Vibration in Women With Phonotraumatic and Nonphonotraumatic Vocal Hyperfunction.

Purpose The purpose of this study was to determine whether estimates of glottal aerodynamic measures based on neck-surface vibration are comparable to those previously obtained using oral airflow and air pressure signals (Espinoza et al., 2017) in terms of discriminating patients with phonotraumatic and nonphonotraumatic vocal hyperfunction (PVH and NPVH) from vocally healthy controls. Method Consecutive /pae/ syllables at comfortable and loud level were produced by 16 women with PVH (organic vocal fold lesions), 16 women with NPVH (primary muscle tension dysphonia), and 32 vocally healthy women who were each matched to a patient according to age and occupation. Subglottal impedance-based inverse filtering of the anterior neck-surface accelerometer (ACC) signal yielded estimates of peak-to-peak glottal airflow, open quotient, and maximum flow declination rate. Average subglottal pressure and microphone-based sound pressure level (SPL) were also estimated from the ACC signal using subject-specific linear regression models. The ACC-based measures of glottal aerodynamics were normalized for SPL and statistically compared between each patient and matched-control group. Results Patients with PVH and NPVH exhibited lower SPL-normalized glottal aerodynamics values than their respective control subjects (p values ranging from < .01 to .07) with very large effect sizes (1.04-2.16), regardless of loudness condition or measurement method (i.e., ACC-based values maintained discriminatory power). Conclusions The results of this study demonstrate that ACC-based estimates of most glottal aerodynamic measures are comparable to those previously obtained from oral airflow and air pressure (Espinoza et al., 2017) in terms of differentiating between hyperfunctional (PVH and NPVH) and normal vocal function. ACC-based estimates of glottal aerodynamic measures may be used to assess vocal function during continuous speech and enables this assessment of daily voice use during ambulatory monitoring to provide better insight into the pathophysiological mechanisms associated with vocal hyperfunction.

CMFALL: A Cascade and Parallel Multi-State Fall Detection Algorithm Using Waist-Mounted Tri-Axial Accelerometer Signals

As one of the main threats to people's health, especially for the elderly, falls have caused a large number of accidents. Detecting falls in time can minimize the severity of the injury and save lives. Therefore, designing fall detection algorithms to assist people's daily life has been a hot research topic. In the last decade, different fall detection approaches based on wearable sensors have been proposed. However, since the hardware resources of wearable sensors are very limited, designing accurate but energy-efficient fall detection algorithms remains an open challenge. To deal with this, an accurate but low-cost fall detection algorithm is proposed in this paper. Particularly, a novel cascade and parallel method that efficiently employs the characteristics of human falls and the advanced modeling ability of the neural network is proposed. Also, a novel design of a lightweight convolutional neural network with self-attention is proposed to achieve the best recognition/numerical complexity tradeoff. The proposed method, namely CMFALL, is evaluated together with a multitude of state-of-the-art models on a large dataset, where it performs the best with an F1-score exceeding 99%. Meanwhile, its computational cost and model size are only a few thousandths of other models. Such low computational cost and small size not only enable to embed it in a wearable sensor but also make the system power requirements quite low, which can enhance the autonomy of the wearable fall detector.

RETRACTED ARTICLE: Application of back propagation artificial neural network in detection and analysis of diabetes mellitus

Diabetes Mellitus affects adults and children, causing changes in lifestyle. The diabetic affected person count has increased drastically worldwide over the last few years; about 425 million people have diabetes. By 2030, it is predicted that diabetic disorder will be the seventh leading cause of human death. Diabetes mellitus is measured invasively. This method has limitations such as patient’s preparation, piercing of the skin, which causes infection and needs for skilled technicians. In order to avoid the limitations of invasive methods, vibrations from the pancreas are acquired using a smartphone accelerometer sensor and detecting the value of diabetes. The human body has a unique energy signature for every organ, which leads to vibrations with different frequencies. The frequency of the vibration signal from the pancreas is proportional to insulin secretion and dynamics. The signals obtained from the accelerometer sensor are trained and analyzed with the Levenberg–Marquardt algorithm for obtaining the relation between the excess insulin secretion and clinical value of the diabetic level of the person. The accelerometer signals and clinical values are modeled with Regression analysis for the diabetic and non-diabetic persons. The results show the correlation between the fluid dynamics of insulin and clinical value at about 95% in prediction.

A Novel Functional Link Network Stacking Ensemble with Fractal Features for Multichannel Fall Detection

Falls are a major health concern and result in high morbidity and mortality rates in older adults with high costs to health services. Automatic fall classification and detection systems can provide early detection of falls and timely medical aid. This paper proposes a novel Random Vector Functional Link (RVFL) stacking ensemble classifier with fractal features for classification of falls. The fractal Hurst exponent is used as a representative of fractal dimensionality for capturing irregularity of accelerometer signals for falls and other activities of daily life. The generalised Hurst exponents along with wavelet transform coefficients are leveraged as input feature space for a novel stacking ensemble of RVFLs composed with an RVFL neural network meta-learner. Novel fast selection criteria are presented for base classifiers founded on the proposed diversity indicator, obtained from the overall performance values during the training phase. The proposed features and the stacking ensemble provide the highest classification accuracy of 95.71% compared with other machine learning techniques, such as Random Forest (RF), Artificial Neural Network (ANN) and Support Vector Machine. The proposed ensemble classifier is 2.3× faster than a single Decision Tree and achieves the highest speedup in training time of 317.7× and 198.56× compared with a highly optimised ANN and RF ensemble, respectively. The significant improvements in training times of the order of 100× and high accuracy demonstrate that the proposed RVFL ensemble is a prime candidate for real-time, embedded wearable device–based fall detection systems.

Your Knock Is My Command: Binary Hand Gesture Recognition on Smartphone with Accelerometer

Motion-based hand gesture is an important scheme to allow users to invoke commands on their smartphones in an eyes-free manner. However, the existing scheme is facing some problems. On the one hand, the expression ability of one single gesture is limited. As a result, a gesture set consisting of multiple gestures is typically adopted to represent different commands. Users must memorize all gestures in order to make interaction successfully. On the other hand, the design of gestures needs to be complicated to express diverse intensions. However, complex gestures are difficult to learn and remember. In addition, complex gestures set a high recognition barrier to smart APPs. This leads to an imbalance problem. Different gestures have different recognition accuracy levels, which may result in instability of recognition precision in practical applications. To address these problems, this paper proposes a novel scheme using binary motion gestures. Only two simple gestures are required to express bit “0” and “1,” and rich information can be expressed through the permutation and combination of the two binary gestures. Firstly, four kinds of candidate binary gestures are evaluated for eyes-free interactions. Then, an online signal cutting and merging algorithm is designed to split accelerometer signals sequence into multiple separate gesture signal segments. Next, five algorithms, including Dynamic Time Warping (DTW), Naive Bayes, Decision Tree, Support Vector Machine (SVM), and Bidirectional Long Short-Term Memory (BLSTM) Network, are adopted to recognize these segments of knock gestures. The BLSTM achieves the top performance in terms of both recognition accuracy and recognition imbalance. Finally, an Android application is developed to illustrate the usability of the proposed binary gestures. As binary gestures are much simpler than traditional hand gestures, they are more efficient and user-friendly. Our scheme eliminates the imbalance problem and achieves high recognition accuracy.

Condition Monitoring of a Paper Feeding Mechanism Using Model-of-Signals as Machine Learning Features

Prognostics and Health Management of machine devices and parts is a hot topic in the Industry 4.0 era. In this fashion, automated procedures to evaluate machinery working conditions are essential to minimize downtime and maintenance costs. In this work, we study how to monitor the decrease in performance of a paper sheet feeder for the packaging industry under heavy-duty cycle operations. The main measurable outcome of such degradation is the increase in backlash among the device moving components. A wide variety of methods and procedures is available to tackle this monitoring problem. In this paper, we analyze the use of a simple yet efficient diagnosis methodology that can exploit machinery controllers (i.e., Programmable Logic Controllers) edge-computing capabilities. Vibration measurements are known in the literature to retain information about the system's mechanics. Model-of Signals, a data-driven approach based on black box system identification, allows to extract that information reliably during machinery working cycle. The refinement of those data using machine learning allows the retrieval of knowledge about the health state of the machine. In this study, the feeder mechanism is run to failure with its parts backlash measured at given time intervals. Accelerometer signals are modelled as AutoRegressive processes whose coefficients are then considered as features to feed to machine learning algorithms, which are employed to perform severity evaluation of the ongoing degradation. Estimation and prediction are both implementable on-board the controller, while the learning task can be carried out remotely, in a cloud computing perspective. The exploitation of AutoRegressive modelling gives a simple and inherent methodology for feature selection, serving as a foundation of the machine learning stage. We make use of a Support Vector Machine algorithm to analyze how obtained models represent the various levels of backlash in the device and develop a suitable predictor of the degradation severity. Finally, the results of the application of the methodology to the case study are shown.

Driver Identification and Verification From Smartphone Accelerometers Using Deep Neural Networks

This paper addresses driver identification and verification using Deep Learning (DL) on tri-axial accelerometer signals from drivers’ smartphones. The proposed driver identification architecture includes ResNet-50 followed by two Stacked Gated Recurrent Units (SGRUs). ResNet provides a deep layer model, thanks to shortcut connections, is able to extract rich features from accelerometers, and GRU layers model the dynamics of drivers’ behavior. ResNet-50 pre-trained on image classification has been evaluated testing two approaches to map 1D accelerometer signals into 2D images. Siamese Neural Networks and Triplet Loss Training have been proposed for driver verification. The Siamese architecture is built on the same ResNet-50 + GRU model of driver identification, while the Triplet loss has required obtaining embeddings at journey level. Experimental results have been obtained for a dataset of 25 drivers, performing 20,025 daily life journeys with more than 800 per driver. Driver identification achieved top-1 and top-5 accuracies of 71.89% and 92.02%, respectively, and driver verification a F1 score of 74.09%. These results are competitive with state-of-the-art research that have generally tested smaller databases (in many cases based only on predefined routes), and have relied on information sources other than accelerometers, such as gyroscopes, magnetometers and GPS. Therefore, we believe that the proposed DL architectures are suitable for developing efficient driver monitoring applications based on only energy-efficient smartphone accelerometer signals.

Human body activity recognition using wearable inertial sensors integrated with a feature extraction–based machine-learning classification algorithm

This study employed wearable inertial sensors integrated with an activity-recognition algorithm to recognize six types of daily activities performed by humans, namely walking, ascending stairs, descending stairs, sitting, standing, and lying. The sensor system consisted of a microcontroller, a three-axis accelerometer, and a three-axis gyro; the algorithm involved collecting and normalizing the activity signals. To simplify the calculation process and to maximize the recognition accuracy, the data were preprocessed through linear discriminant analysis; this reduced their dimensionality and captured their features, thereby reducing the feature space of the accelerometer and gyro signals; they were then verified through the use of six classification algorithms. The new contribution is that after feature extraction, data classification results indicated that an artificial neural network was the most stable and effective of the six algorithms. In the experiment, 20 participants equipped the wearable sensors on their waists to record the aforementioned six types of daily activities and to verify the effectiveness of the sensors. According to the cross-validation results, the combination of linear discriminant analysis and an artificial neural network was the most stable classification algorithm for data generalization; its activity-recognition accuracy was 87.37% on the training data and 80.96% on the test data.

Fault Diagnosis of Rotary Machines Using Deep Convolutional Neural Network with Wide Three Axis Vibration Signal Input.

Fault diagnosis is considered as an essential task in rotary machinery as possibility of an early detection and diagnosis of the faulty condition can save both time and money. This work presents developed and novel technique for deep-learning-based data-driven fault diagnosis for rotary machinery. The proposed technique input raw three axes accelerometer signal as high definition 1D image into deep learning layers which automatically extract signal features, enabling high classification accuracy. Unlike the researches carried out by other researchers, accelerometer data matrix with dimensions 6400 × 1 × 3 is used as input for convolutional neural network training. Since convolutional neural networks can recognize patterns across input matrix, it is expected that wide input matrix containing vibration data should yield good classification performance. Using convolutional neural networks (CNN) trained model, classification in one of the four classes can be performed. Additionally, number of kernels of CNN is optimized using grid search, as preliminary studies show that alternating number of kernels impacts classification results. This study accomplished the effective classification of different rotary machinery states using convolutional artificial neural network for classification of raw three axis accelerometer signal input.

Diagnosis of Hand Tremor from EMG Sensors

Parkinson’s disease (PD) initially affects on motor activities of the patient. The symptoms of the disease include tremors, muscular rigidity, changes in gait, and complications in speech. In this paper, the experimental procedure for the acquisition of EMG and accelerometer data is explained. From the data, tremor frequency is evaluated. Tremor is an involuntary shaking of the body or the limbs which is considered as an early symptom of Parkinson’s disease. EMG instrument, sensors, selection of muscles in the study, data acquisition procedure, analysis of EMG and accelerometer signal are discussed in this paper. Then the signals were analyzed in the frequency domain to assess tremor frequency in the hand. Since the signals were taken from the underlying muscles of the skin, the results are relatively accurate than indirect methods of hand tremor assessment. Cite as: Mousa M. AL-Mutari, K. Prahlad Rao, Ibrahim M. Mehedi (2020). Diagnosis of Hand Tremor from EMG Sensors. Journal of Technological Science & Engineering (JTSE), 01(01), 30-35. http://doi.org/10.5281/zenodo.3940857

Influences of Sensor Placement Site and Subject Posture on Measurement of Respiratory Frequency Using Triaxial Accelerometers.

IntroductionRespiration frequency (RF) could be derived from the respiratory signals recorded by accelerometers which detect chest wall movements. The optimum direction of acceleration for accurate RF measurement is still uncertain. We aim to investigate the effect of measure site, posture, and direction of acceleration on the accuracy of accelerometer-based RF estimation.MethodsIn supine and seated postures respectively, respiratory signals were measured from 34 healthy subjects in 70 s by triaxial accelerometers located at four sites on the body wall (over the clavicle, laterally on the chest wall, over the pectoral part of the anterior chest wall, on the abdomen in the midline at the umbilicus), with the reference respiratory signal simultaneously recorded by a strain gauge chest belt. RFs were extracted from the accelerometer and reference respiratory signals using wavelet transformation. To investigate the effect of measure site, posture, and direction of acceleration on the accuracy of accelerometer-based RF estimation, repeated measures multivariate analysis of variance, linear regression, Bland-Altman analysis, and Scheirer-Ray-Hare test were performed between reference and accelerometer-based RFs.ResultsThere was no significant difference in accelerometer-based RF estimation between seated and supine postures, among four accelerometer sites, or between seated or supine postures (p > 0.05 for all). The error of accelerometer-based RF estimation was less than 0.03 Hz (two breaths per minute) at any site or posture, but was significantly smaller in supine posture than in seated posture (p < 0.05), with narrower limits of agreement in Bland-Altman analysis and higher accuracy in linear regression (R2 > 0.61 vs. R2 < 0.51).ConclusionRespiration frequency can be accurately measured from the acceleration of any direction using triaxial accelerometers placed at the clavicular, pectoral and lateral sites on the chest as well the mid abdominal site. More accurate RF estimation could be achieved in supine posture compared with seated posture.

Automated Relative Fundamental Frequency Algorithms for Use With Neck-Surface Accelerometer Signals

Relative fundamental frequency (RFF) has been suggested as a potential acoustic measure of vocal effort. However, current clinical standards for RFF measures require time-consuming manual markings. Previous semi-automated algorithms have been developed to calculate RFF from microphone signals. The current study aimed to develop fully automated algorithms to calculate RFF from neck-surface accelerometer signals for ecological momentary assessment and ambulatory monitoring of voice. Training a set of 2646 /vowel-fricative-vowel/ utterances from 317 unique speakers, with and without voice disorders, was used to develop automated algorithms to calculate RFF values from neck-surface accelerometer signals. The algorithms first rejected utterances with poor vowel-to-noise ratios, then identified fricative locations, then used signal features to determine voicing boundary cycles, and finally calculated corresponding RFF values. These automated RFF values were compared to the clinical gold-standard of manual RFF calculated from simultaneously collected microphone signals in a novel test set of 639 utterances from 77 unique speakers. Automated accelerometer-based RFF values resulted in an average mean bias error (MBE) across all cycles of 0.027 ST, with an MBE of 0.152 ST and -0.252 ST in the offset and onset cycles closest to the fricative, respectively. All MBE values were smaller than the expected changes in RFF values following successful voice therapy, suggesting that the current algorithms could be used for ecological momentary assessment and ambulatory monitoring via neck-surface accelerometer signals.

Vibration Generator Device Based on Industrial Vibrators

A low-cost vibration generator device based on industrial vibrators was designed. The control software was implemented in LabVIEW Environment. The device is able to generate an oscillating force of 8 kN and an amplitude of up to 4 mm at a frequency of 50 Hz to model low-amplitude, high-frequency vehicle vibrations. A National Instruments myRIO device was responsible for data acquisition, with which a signal of a piezoelectric accelerometer was detected. The test results show that the device is able to generate a sinusoidal harmonic acceleration.

Freezing of gait detection in Parkinson's disease via multimodal analysis of EEG and accelerometer signals.

Parkinson's disease (PD) patients with freezing of gait (FOG) can suddenly lose their forward moving ability leading to unexpected falls. To overcome FOG and avoid the falls, a real-time accurate FOG detection or prediction system is desirable to trigger on-demand cues. In this study, we designed and implemented an in-place movement experiment for PD patients to provoke FOG and meanwhile acquired multimodal physiological signals, such as electroencephalography (EEG) and accelerometer signals. A multimodal model using brain activity from EEG and motion data from accelerometers was developed to improve FOG detection performance. In the detection of over 700 FOG episodes observed in the experiments, the multimodal model achieved 0.211 measured by Matthews Correlation Coefficient (MCC) compared with the single-modal models (0.127 or 0.139).Clinical Relevance- This is the first study to use multimodal: EEG and accelerometer signal analysis in FOG detection, and an improvement was achieved.

Derivative Based Gait Event Detection Algorithm Using Unfiltered Accelerometer Signals.

Wearable sensors have been investigated for the purpose of gait analysis, namely gait event detection. Many types of algorithms have been developed specifically using inertial sensor data for detecting gait events. Though much attention has turned toward machine learning algorithms, most of these approaches suffer from large computational requirements and are not yet suitable for real-time applications such as in prostheses or for feedback control. Current rules-based algorithms for real-time use often require fusion of multiple sensor signals to achieve high accuracy, thus increasing complexity and decreasing usability of the instrument. We present our results of a novel, rules-based algorithm using a single accelerometer signal from the foot to reliably detect heel-strike and toe-off events. Using the derivative of the raw accelerometer signal and applying an optimizer and windowing approach, high performance was achieved with a sensitivity and specificity of 94.32% and 94.70% respectively, and a timing error of 6.52 ± 22.37 ms, including trials involving multiple speed transitions. This would enable development of a compact wearable system for robust gait analysis in real-world settings, providing key insights into gait quality with the capability for real-time system control.

Robust PPG-based Ambulatory Heart Rate Tracking Algorithm.

Recent advances in wearable devices with optical Photoplethysmography (PPG) and actigraphy have enabled inexpensive, accessible, and convenient Heart Rate (HR) monitoring. Nevertheless, PPG's susceptibility to motion presents challenges in obtaining reliable and accurate HR estimates during ambulatory and intense activity conditions. This study proposes a lightweight HR algorithm, TAPIR: a Time-domain based method involving Adaptive filtering, Peak detection, Interval tracking, and Refinement, using simultaneously acquired PPG and accelerometer signals. The proposed method is applied to four unique, wrist-wearable based, publicly available databases that capture a variety of controlled and uncontrolled daily life activities, stress, and emotion. The results suggest that the current HR prediction is significantly (P<0.01) more accurate during intense activity conditions than the contemporary algorithms involving Wiener filtering, time-frequency analysis, and deep learning. The current HR tracking algorithm is validated to be of clinical-grade and suitable for low-power embedded wearable systems as a powerful tool for continuous HR monitoring in real-world ambulatory conditions.

Estimation of Subglottal Pressure From Neck Surface Vibration in Patients With Voice Disorders.

Purpose Given the established linear relationship between neck surface vibration magnitude and mean subglottal pressure (Ps) in vocally healthy speakers, the purpose of this study was to better understand the impact of the presence of a voice disorder on this baseline relationship. Method Data were obtained from participants with voice disorders representing a variety of glottal conditions, including phonotraumatic vocal hyperfunction, nonphonotraumatic vocal hyperfunction, and unilateral vocal fold paralysis. Participants were asked to repeat /p/-vowel syllable strings from loud-to-soft loudness levels in multiple vowel contexts (/pa/, /pi/, /pu/) and pitch levels (comfortable, higher than comfortable, lower than comfortable). Three statistical metrics were computed to analyze the regression line between neck surface accelerometer (ACC) signal magnitude and Ps within and across pitch, vowel, and voice disorder category: coefficient of determination (r 2), slope, and intercept. Three linear mixed-effects models were used to evaluate the impact of voice disorder category, pitch level, and vowel context on the relationship between ACC signal magnitude and Ps. Results The relationship between ACC signal magnitude and Ps was statistically different in patients with voice disorders than in vocally healthy controls; patients exhibited higher levels of Ps given similar values of ACC signal magnitude. Negligible effects were found for pitch condition within each voice disorder category, and negligible-to-small effects were found for vowel context. The mean of patient-specific r 2 values was .63, ranging from .13 to .92. Conclusions The baseline, linear relationship between ACC signal magnitude and Ps is affected by the presence of a voice disorder, with the relationship being participant-specific. Further work is needed to improve ACC-based prediction of Ps, across treatment, and during naturalistic speech production.

Comparison of four methods for determining the cut-off frequency of accelerometer signals in able-bodied individuals and ACL ruptured subjects

BackgroundSelecting the appropriate cut-off frequency to filter triaxial accelerometric data is a challenging issue in gait analyses. It reduces soft tissues artifacts and the variability in the kinematic data waveform from able-bodied and physically impaired gaits. Research questionAre cut-off frequencies estimated by four filtering methods similar along each axis of a triaxial accelerometer and for able-bodied subjects and those with an anterior cruciate ligament rupture (ACLR)? MethodsAfter walking on a treadmill, the cut-off frequency for the tibial accelerations was calculated using 95 and 99 per cent of the energy spectrum (E), residual analysis (RA) and, the method (Yu) proposed by Yu et al. [1]. The coefficient of variation was used to express the variability of the cut-off frequencies estimated by the four methods. t-Test and repeated measure ANOVA were applied to examine the effects of healthiness and acceleration axis on cut-off frequencies. ResultsOn average, E95 and E99 gave the lowest and the highest cut-off frequencies respectively. The results demonstrated the effect of ACL injury and axes on the cut-off frequencies, especially on the RA method. There was a significant difference in the cut-off frequencies between healthy and ACLR subjects for the vertical axis with the RA method and for the anterior-posterior (AP) axis with the Yu method. Similar cut-off frequencies were obtained for all axes with the E99 method for within groups’ comparison. The E95 and E99 methods gave the least and most variable outputs respectively. Significant within group differences between cut off frequencies calculated by four methods, led to disappearing peaks in the more fluctuating portion of the acceleration data. SignificanceA single cut-off frequency is not recommended for all individuals and axes. In cases where a single cut-off frequency is necessary for all individuals or axes, RA or E99 methods are suggested respectively.