Oxygen Saturation Variability: Healthy Adults

A New Way of Data Analysis and Rating of the Blood Oxygen Saturation Variability

“…The records of SpO2 have a specific look [6], [7], thereby hint out their multi-fractal nature. Here we are going to confirm the results [8] by the independent way, basing on the statistical frequencies plots and kernel probability density estimations (KDE, [9], [10]). KDE has been worked out for the continuous probability distributions.…”

“…The data set has been got in [7], submitted in the Physionet portal [11], and investigated previously either partly [8] or integrally [6], [7]. This set involves the SpO2 one-hour records with one Hertz sampling rate for 36 healthy adults.…”

“…Prior works cited above have pointed out some hallmarks of the data. Those were non-Gauss distributions with massive outliers [6], [8], hard clusterization for Poincare Plots with the prevailing of the long-time variability [7], [8], and the close negative correlation among the Reccurece Ratios and variability descriptors [6].…”

“…The variability research for peripheral blood oxygenation (SpO2) meets the principal difficulties [6]- [8]. Primary time series (records [6]) are non-Gaussian in essence [6], [8].…”

“…The variability research for peripheral blood oxygenation (SpO2) meets the principal difficulties [6]- [8]. Primary time series (records [6]) are non-Gaussian in essence [6], [8]. That at once stays under doubt usual statistical measures as standard deviations (STD), interquartile ranges (IQR), and ranges (Rs) of the samples.…”