Heart rate variability analysis using different wavelet transformations

Heart rhythm is modulated by the autonomic nervous system in order to adapt to changing cardiovascular loads. The resulting heart rate variability (HRV) is considered as one of the most promising non-invasive markers of the activity of the autonomic nervous system. In this work, the possibility to analyze HRV with wavelet transform was investigated. Specifically, three methods of wavelet transform (CWT, CWTFT and DWT) were compared. DWT showed to be the best one because of the simplicity of decomposition, and quality of reconstruction, high time resolution as well as the shortest computation time and lowest CPU power consumption. DWT showed to have high noise robustness, excellent temporal and good frequency resolution and small influence of the mother wavelet type. ULF, HF and LF components of HRV were assessed using DWT transform. LF/HF ratio was extracted from HRV signal during voluntary breathing apneas and compared to the respiration signal. An increase in LF/HF ratio corresponding to the times of breathing apnea could be observed, indicating that the proposed method could reveal specific changes in the HRV and that LF/HF ratio could serve as a good indicator of breathing apnea.