Time Series Of Human Gait Research Articles

Analyzing gait symmetry with automatically synchronized wearable sensors in daily life

Gait deviations such as asymmetry are one of the characteristic symptoms of motor dysfunctions that contribute to the risk of falls. Our objective is to measure gait abnormalities such as asymmetry of the lower limbs in order to evaluate the diagnosis more objectively. For the measurement we use inertial measurement unit (IMU) sensors and force sensors, which are integrated in wristbands and insoles. To extend the battery life of wearable devices, we only save data of the activity gait within the wearables. Therefore we perform activity recognition with a smartphone. Using convolutional neural network (CNN) we achieved an accuracy of 94.7 % of the activity gait recognition. Before recording we synchronize the wearable sensors and reach a maximum latencies of 3 ms. Before the analysis of the symmetry we detect the strides by using a CNN with an accuracy of 98.8 %. For the symmetry evaluation we used dynamic time warping (DTW). The DTW enables us to calculate symmetry of the complete time series of human gait.

Multivariate Multiscale Symbolic Entropy Analysis of Human Gait Signals

The complexity quantification of human gait time series has received considerable interest for wearable healthcare. Symbolic entropy is one of the most prevalent algorithms used to measure the complexity of a time series, but it fails to account for the multiple time scales and multi-channel statistical dependence inherent in such time series. To overcome this problem, multivariate multiscale symbolic entropy is proposed in this paper to distinguish the complexity of human gait signals in health and disease. The embedding dimension, time delay and quantization levels are appropriately designed to construct similarity of signals for calculating complexity of human gait. The proposed method can accurately detect healthy and pathologic group from realistic multivariate human gait time series on multiple scales. It strongly supports wearable healthcare with simplicity, robustness, and fast computation.

Extracting Fundamental Periods to Segment Biomedical Signals.

We address the problem of segmenting nearly periodic time series into period-like segments. We introduce a definition of nearly periodic time series via triplets 〈 basic shape, shape transformation, time scaling 〉 that covers a wide range of time series. To split the time series into periods, we select a pair of principal components of the Hankel matrix. We then cut the trajectory of the selected principal components by its symmetry axis and, thus, obtaining half-periods that are merged into segments. We describe a method of automatic selection of periodic pairs of principal components, corresponding to the fundamental periodicity. We demonstrate the application of the proposed method to the problem of period extraction for accelerometric time series of human gait. We see the automatic segmentation into periods as a problem of major importance for human activity recognition problem, since it allows to obtain interpretable segments: each extracted period can be seen as an ultimate entity of gait. The method we propose is more general compared to the application specific methods and can be used for any nearly periodical time series. We compare its performance to classical mathematical methods of period extraction and find that it is not only comparable to the alternatives, but in some cases performs better.

Symbolic Time Series Analysis of Temporal Gait Dynamics

Signals obtained from biological systems exhibit pronounced complexity. The patterns of change contain valuable information about the dynamics of underlying control mechanism of the complex biological systems. Human gait is a complex process with multiple inputs and numerous outputs. Various complexity analysis tools have been proposed to extract information from human gait time series. In this study, we used recently developed threshold based symbolic entropy to compare the spontaneous output of the human locomotors system during constrained and metronomically paced walking protocols. For that purpose, stride interval time series of healthy subjects who walked for 1 h at normal, slow and fast rates under different conditions was transformed into symbol sequences. Normalized corrected Shannon entropy (NCSE) was computed from the symbol sequences of the stride interval time series. The findings indicated that the unprompted output of human locomotors system is more complex during unconstrained normal walking as compared with slow, fast or metronomically paced walking.

Multifractal detrended fluctuation analysis of human gait diseases

In this paper multifractal detrended fluctuation analysis (MFDFA) is used to study the human gait time series for normal and diseased sets. It is observed that long range correlation is primarily responsible for the origin of multifractality. The study reveals that the degree of multifractality is more for normal set compared to diseased set. However, the method fails to distinguish between the two diseased sets.

Effect of treadmill walking on the stride interval dynamics of human gait

Metronomic walking has been found to diminish the statistical persistence intrinsic to the stride interval time series of human gait. Since treadmill walking (TW) possesses a similar form of external pacing, we proposed to study the disruptions in the natural neuromuscular rhythms of gait during TW. Treadmill walking is a widespread rehabilitative tool, however, its effect on an individual’s stride dynamics is not well understood. To better elucidate potential effects, we tested the hypothesis that TW without handrails would diminish the statistical persistence in an individual’s stride interval time series. The scaling exponent ( α ) was employed in this study as a measure of the statistical persistence of the stride interval time series. Sixteen able-bodied young adults (mean age: 23.3 ± 3.3 years) were instructed to walk at a self-selected comfortable pace for 15 min in three different conditions in a randomized order: (1) overground walking, (2) TW without holding a handrail (NoRail) and (3) TW while holding a front handrail (Rail). The α did not differ significantly between the overground and NoRail conditions ( P > 0.5 ). However, the α of the Rail condition ( α = 0.92 ± 0.10 ) differed significantly from both the overground ( α = 0.83 ± 0.06 ; P < 0.015 ) and NoRail conditions ( α = 0.82 ± 0.08 ; P < 0.01 ). In contrast, stride interval variability did not change between walking conditions ( P > 0.5 ) . These findings indicate that comfortable-paced TW does not diminish the intrinsic stride dynamics of human gait.

A non linear analysis of human gait time series based on multifractal analysis and cross correlations

We analyzed databases with gait time series of adults and persons with Parkinson, Huntington and amyotrophic lateral sclerosis (ALS) diseases. We obtained the staircase graphs of accumulated events that can be bounded by a straight line whose slope can be used to distinguish between gait time series from healthy and ill persons. The global Hurst exponent of these series do not show tendencies, we intend that this is because some gait time series have monofractal behavior and others have multifractal behavior so they cannot be characterized with a single Hurst exponent. We calculated the multifractal spectra, obtained the spectra width and found that the spectra of the healthy young persons are almost monofractal. The spectra of ill persons are wider than the spectra of healthy persons. In opposition to the interbeat time series where the pathology implies loss of multifractality, in the gait time series the multifractal behavior emerges with the pathology. Data were collected from healthy and ill subjects as they walked in a roughly circular path and they have sensors in both feet, so we have one time series for the left foot and other for the right foot. First, we analyzed these time series separately, and then we compared both results, with direct comparison and with a cross correlation analysis. We tried to find differences in both time series that can be used as indicators of equilibrium problems.

Multiscale entropy analysis of human gait dynamics

We compare the complexity of human gait time series from healthy subjects under different conditions. Using the recently developed multiscale entropy algorithm, which provides a way to measure complexity over a range of scales, we observe that normal spontaneous walking has the highest complexity when compared to slow and fast walking and also to walking paced by a metronome. These findings have implications for modeling locomotor control and for quantifying gait dynamics in physiologic and pathologic states.

Stratification of the phase clouds and statistical effects of the non-Markovity in chaotic time series of human gait for healthy people and Parkinson patients

In this work we develop a new method of diagnosing the nervous system diseases and a new approach in studying human gait dynamics with the help of the theory of discrete non-Markov random processes (Phys. Rev. E 62 (5) (2000) 6178, Phys. Rev. E 64 (2001) 066132, Phys. Rev. E 65 (2002) 046107, Physica A 303 (2002) 427). The stratification of the phase clouds and the statistical non-Markov effects in the time series of the dynamics of human gait are considered. We carried out the comparative analysis of the data of four age groups of healthy people: children (from 3 to 10 year olds), teenagers (from 11 to 14 year olds), young people (from 21 up to 29 year olds), elderly persons (from 71 to 77 year olds) and Parkinson patients. The full data set are analyzed with the help of the phase portraits of the four dynamic variables, the power spectra of the initial time correlation function and the memory functions of junior orders, the three first points in the spectra of the statistical non-Markov parameter. The received results allow to define the predisposition of the probationers to deflections in the central nervous system caused by Parkinson's disease. We have found out distinct differences between the five submitted groups. On this basis we offer a new method of diagnostics and forecasting Parkinson's disease.