Physical conditioning status stratification based on heart rate variability: Principal component analysis of power spectrum density function

The purpose was to investigate the application of principal components analysis (PCA) on power spectral density function (PSD) computed by heart rate variability (HRV) signals aiming at distinguishing physical conditioning status. Ten healthy sedentary volunteers (controls) and ten matched professional long distance runners (athlete) were enrolled. The subjects had their maximal metabolic equivalents (MET) estimated and 20 min resting ECG recorded. In RR intervals (iRR) series, a 5 min segment with the least variance was selected. iRR sequence was interpolated at 1 Hz (cubic spline) to estimate the PSD by fast Fourier transform. PCA was applied to PSD, and the first three principal components were retained (90% of the total variability). A logistic regression model based on Mahalanobis distance determined the optimal separation threshold. The area under the ROC curve (AUC) of the proposed method was compared to standard LF and HF parameters (α=0.05). The proposed method allows to correctly separate all subjects (AUC=1.0), while the HRV parameters in the frequency domain LF and HF provided AUC equal to 0.8 and 0.9 (p>0.05), respectively. The PCA applied to PSD of iRR series properly classified the physical conditioning of controls and athletes with superior performance to standard HRV.