Improving the understanding of sleep apnea characterization using Recurrence Quantification Analysis by defining overall acceptable values for the dimensionality of the system, the delay, and the distance threshold

Abstract

Our contribution focuses on the characterization of sleep apnea from a cardiac rate point of view, using Recurrence Quantification Analysis (RQA), based on a Heart Rate Variability (HRV) feature selection process. Three parameters are crucial in RQA: those related to the embedding process (dimension and delay) and the threshold distance. There are no overall accepted parameters for the study of HRV using RQA in sleep apnea. We focus on finding an overall acceptable combination, sweeping a range of values for each of them simultaneously. Together with the commonly used RQA measures, we include features related to recurrence times, and features originating in the complex network theory. To the best of our knowledge, no author has used them all for sleep apnea previously. The best performing feature subset is entered into a Linear Discriminant classifier. The best results in the “Apnea-ECG Physionet database” and the “HuGCDN2014 database” are, according to the area under the receiver operating characteristic curve, 0.93 (Accuracy: 86.33%) and 0.86 (Accuracy: 84.18%), respectively. Our system outperforms, using a relatively small set of features, previously existing studies in the context of sleep apnea. We conclude that working with dimensions around 7–8 and delays about 4–5, and using for the threshold distance the Fixed Amount of Nearest Neighbours (FAN) method with 5% of neighbours, yield the best results. Therefore, we would recommend these reference values for future work when applying RQA to the analysis of HRV in sleep apnea. We also conclude that, together with the commonly used vertical and diagonal RQA measures, there are newly used features that contribute valuable information for apnea minutes discrimination. Therefore, they are especially interesting for characterization purposes. Using two different databases supports that the conclusions reached are potentially generalizable, and are not limited by database variability.

Recurrence Quantitative Analysis of Wavelet-Based Surrogate Data for Nonlinearity Testing in Heart Rate Variability

“…To study the behavior of nonstationary signals, the fixed amount of nearest neighbors (FAN), in which the radius ε changes for each point, leads to an asymmetric recurrence plot in which all columns have the same recurrence density despite the nonstationary behavior, or trends, in the time series ( Marwan, 2011 ). Therefore, to address this phenomenon, FAN norm has been recommended for analyzing HRV ( Martín-González et al, 2018 ).…”

“…The analysis of nonlinearity in these settings has led to the increasing application of nonlinear tools in which time series are not required be neither very long nor nonstationary; this is the case of the recurrence plots (RP) ( Marwan et al, 2002 , 2007 ). Recurrence quantitative analysis (RQA) is used to quantify diverse nonlinear behaviors in the RP and is widely used in physiological time series, such as electroencephalography (EEG) ( Ouyang et al, 2008 ; Heunis et al, 2018 ; Pitsik et al, 2020 ) and ECG ( Marwan et al, 2002 ; Naschitz et al, 2003 ; Gonzalez et al, 2013 ; Martín-González et al, 2018 ).…”

Physiological and Behavioral Factors in Musicians’ Performance Tempo

“…Embedding dimensions were chosen on an individual basis and ranged from 4 to 8. A Theiler window fixed to the time delay ( Javorka et al, 2009 ) was applied to the data, as cardiac signals tend to show high autocorrelation ( Martin-Gonzalez et al, 2018 ). Recurrence rate, the percentage of recurrent points in the system, was fixed to 5% as per previous RQA studies of cardiac signals ( Javorka et al, 2008 , 2009 ).…”

Bispectral Analysis of Heart Rate Variability to Characterize and Help Diagnose Pediatric Sleep Apnea

“…Although HRV signals are usually nonlinear and non-Gaussian in nature [29,30], under some specific conditions, HRV can show dynamics where these nonlinearities are not always present [31]. Nevertheless, in the context of pediatric OSA, the nonlinear dynamics of HRV signals can be increased during sleep [21,32], as well as by cardiac alterations due to OSA [20,21,30,32]. Furthermore, in the present study, the presence of HRV nonlinear dynamics in the pediatric OSA context has been demonstrated (see Appendix C).…”