This exploratory study provides evidence that, despite some redundancies in the informative content of nonlinear indices and strong differences in their prognostic power, quantification of nonlinear properties of HRV provides independent information in risk stratification of CHF patients.

Non-invasive fetal heart rate is of great relevance in clinical practice to monitor fetal health state during pregnancy. To date, however, despite significant advances in the field of electrocardiography, the analysis of abdominal fetal ECG is considered a challenging problem for biomedical and signal processing communities. This is mainly due to the low signal-to-noise ratio of fetal ECG and difficulties in cancellation of maternal QRS complexes, motion and electromyographic artefacts. In this paper we present an efficient unsupervised algorithm for fetal QRS complex detection from abdominal multichannel signal recordings combining ICA and maternal ECG cancelling, which outperforms each single method. The signal is first pre-processed to remove impulsive artefacts, baseline wandering and power line interference. The following steps are then applied: maternal ECG extraction through independent component analysis (ICA); maternal QRS detection; maternal ECG cancelling through weighted singular value decomposition; enhancing of fetal ECG through ICA and fetal QRS detection. We participated in the Physionet/Computing in Cardiology Challenge 2013, obtaining the top official scores of the challenge (among 53 teams of participants) of event 1 and event 2 concerning fetal heart rate and fetal interbeat intervals estimation section. The developed algorithms are released as open-source on the Physionet website.

Heart rate variability (HRV) is a well-known phenomenon whose characteristics are of great clinical relevance in pathophysiologic investigations. In particular, respiration is a powerful modulator of HRV contributing to the oscillations at highest frequency. Like almost all natural phenomena, HRV is the result of many nonlinearly interacting processes; therefore any linear analysis has the potential risk of underestimating, or even missing, a great amount of information content. Recently the technique of Empirical Mode Decomposition (EMD) has been proposed as a new tool for the analysis of nonlinear and nonstationary data. We applied EMD analysis to decompose the heartbeat intervals series, derived from one electrocardiographic (ECG) signal of 13 subjects, into their components in order to identify the modes associated with breathing. After each decomposition the mode showing the highest frequency and the corresponding respiratory signal were Hilbert transformed and the instantaneous phases extracted were then compared. The results obtained indicate a synchronization of order 1:1 between the two series proving the existence of phase and frequency coupling between the component associated with breathing and the respiratory signal itself in all subjects.

Several parameters assessing nonlinear properties of heart rate variability (HRV) from short-term (<10 min) laboratory recordings have been proposed so far, but their reliability is unknown. In this study, we addressed this issue analysing a comprehensive set of these indices. In 42 healthy subjects (mean age (min-max): 38 (26-56) years, 21 men) we recorded 5 min of supine ECG in two consecutive days. From RR intervals we computed 11 nonlinear HRV indices, representative of symbolic dynamics, entropy, fractality, predictability, empirical mode decomposition and Poincaré plot families. Absolute reliability was assessed by the 95% limits of random variation and relative reliability was assessed computing the intraclass correlation coefficient (ICC). We found marked differences in the reliability of short-term nonlinear indices of HRV. In the majority of indices, changes in test-retest measurements ranged between about -30% and +50%, indicating good absolute reliability while in the others the change was <-60% and >140%. Relative reliability was substantial (0.6 < ICC < 0.8) in half of the indices, moderate in one and poor in the remaining. Compared to classical linear indices, nonlinear HRV parameters seem more suitable for individual test-retest evaluations but, due to a reduced ICC, they need increased sample size in comparative studies involving two groups of subjects.