In the measurement of bio signals associated with the heart rate, artifacts in the electrocardiogram (ECG) recordings deteriorate the data, yielding ECG artifacts; missing (incomplete) RR interval tachogram. The linear parameters of heart rate variability (HRV) are very sensitive to these missing RR intervals. In this study, the feat of nonlinear measures of HRV is investigated for missing RR interval data, using simulated missing data in real RR interval tachograms. For the simulation, randomly selected data (0–100 RR intervals) were removed from real RR data obtained from the MIT-BIH normal sinus rhythm database. all, 703 tachograms of 1000 RR interval data length were used for this analysis in Approximate entropy (ApEn), sample entropy (SampEn), Poincaré plot indices (SD1 and SD2) and Detrended fluctuation analysis (DFA) were calculated as the nonlinear parameters, and the relative errors between the original and the incomplete tachograms for these parameters were computed. The results of the simulation revealed that nonlinear parameters are more suitable measures than linear parameters of HRV in presence of missing RR interval data.
