China
Africa
Explore chapters and articles related to this topic
Historical Development of HRV Analysis
Published in Herbert F. Jelinek, David J. Cornforth, Ahsan H. Khandoker, ECG Time Series Variability Analysis, 2017
The approximate entropy (ApEn) represents a simple index for the overall “complexity” and “predictability” of time series (Pincus 1991). ApEn can be used to determine the degree of irregularity or disorder within a HR time series, measuring the underlying complexity of the system producing the dynamics. ApEn compares runs of patterns in time series; if similar patterns in a HR time series are found, ApEn estimates the logarithmic likelihood that the next intervals after each of the patterns will differ (i.e., the similarity of the patterns is more coincidence and lacks predictive value) (Ho et al. 1997). Two input parameters, m, the length of compared patterns and r, which defines the criterion of similarity, have to be fixed prior to the computation of ApEn. ApEn has revealed good statistically validity for m=2 and r=15% of the standard deviation of the HR time series. If a time series has more regularity and less complexity, the value of ApEn will be small. On the other hand, if a time series has more irregularity and complexity the value of ApEn will be higher. Pincus and Goldberger (1994) suggested that the reduction in entropy during pathology represents the system decoupling from external inputs or a reduction in the influence of these inputs.
Stress emotion classification using optimized convolutional neural network for online transfer learning dataset
Published in Computer Methods in Biomechanics and Biomedical Engineering, 2022
G. Linda Rose, M. Punithavalli
An EEG emotion classification technique (Chen et al. 2020) was proposed according to the LIBSVM classifier. First, EEG features were computed to define the characteristics related to emotional states. Then, the Lempel–Ziv complexity and wavelet detail coefficients were computed for the pre-processed EEG signals. Then, the co-integration correlation was obtained to analyze the association between channels based on the co-integration analysis. Then, EMD was executed on the preprocessed EEG signals and the mean approximate entropy of the initial four IMFs was computed. Further, the computed features were fed to the LIBSVM classifier to achieve sentiment classification of every channel data. Finally, the classification of every channel was combined by the Takagi–Sugeno fuzzy system to obtain the final emotion classification. But, it has a lower recognition rate and needs to classify multi-category emotions.
Exploring the determinants of success in different clusters of ball possession sequences in soccer
Published in Research in Sports Medicine, 2020
Murilo Merlin, Sergio Augusto Cunha, Felipe Arruda Moura, Ricardo da Silva Torres, Bruno Gonçalves, Jaime Sampaio
Forty-one variables were computed and classified into three groups: notational, space occupation, and displacement synchronization (Table 1). Dynamic variables were analysed using the absolute values (mean), normalized approximate entropy (ApEn), and coefficient of variation (CV). ApEn is a nonlinear measure that quantifies the regularity in complex system behaviours (Pincus, 1991). For this study, we decided to compute the normalized entropy, a non-modified measure of regularity derived from the original ApEn, which is less dependent on time series length (Fonseca, Milho, Passos, Araújo, & Davids, 2012). Coefficient of variation (CV) values ((standard deviation/mean)×100) were used to verify the magnitude of variability of the time series.
Effects of pitch spatial references on players’ positioning and physical performances during football small-sided games
Published in Journal of Sports Sciences, 2019
Diogo Coutinho, Bruno Gonçalves, Bruno Travassos, Eduardo Abade, Del P. Wong, Jaime Sampaio
The dynamic positional data of the players were used to determine the team stretch index per minute (Travassos et al., 2014), the distance between teammates’ dyads (n = 15 dyads per team) and the magnitude of the variability in the distance between players’, expressed by the coefficient of variation (CV) (Gonçalves et al., 2017). The approximate entropy (ApEn) technique was applied to analyse the structure of variability expresses in regularity of the time series corresponding to the stretch index and to the distance between players. The ApEn outcome ranges from 0 to 2, in which lower values correspond to more repeatable patterns. For example, values closer to 0 in the distance between dyads, means that the players dyads tend to stay at the same distance in a more repeatable fashion. The imputed values used to compute were 2 to vector length (m) and 0.2*std to the tolerance (r) (Yentes et al., 2013). The ApEn variable was also applied to the players’ zones occupied (Regularity in Zones Occupied). For this purpose, the pitch was divided into 9 zones (formed by the interceptions of pitch lines) and the players’ displacements during the SSG were assigned to the corresponding pitch zone. Then, in the time series corresponding to the movements in the zones occupied by the players was applied the ApEn technique. The intra-team coordination tendencies were assessed based on the time that teammates dyads spent synchronized in both longitudinal and lateral directions. These two last variables were calculated with the Hilbert transform (Palut & Zanone, 2005) and applied for all possible dyads for the six outfield teammates (possible 15 dyads). The near-in-phase synchronization of each dyad was quantified by the percentage of time spent between −30º to 30º bin (near-in-phase mode of coordination) (Folgado et al., 2014).