Multifractal analysis of cardiovascular variability series is an effective tool for the characterization of pathological states associated with congestive heart failure (CHF). Consequently, variations of heartbeat scaling properties have been associated with the dynamical balancing of nonlinear sympathetic/vagal activity. Nevertheless, whether vagal dynamics has multifractal properties yet alone is currently unknown. In this study, we answer this question by conducting multifractal analysis through wavelet leader-based multiscale representations of instantaneous series of vagal activity as estimated from inhomogeneous point process models. Experimental tests were performed on data gathered from 57 CHF patients, aiming to investigate the automatic recognition accuracy in predicting survivor and non-survivor patients after a 4 years follow up. Results clearly indicate that, on both CHF groups, the instantaneous vagal activity displays power-law scaling for a large range of scales, from ≃ 0.5s to ≃ 100s. Using standard SVM algorithms, this information also allows for a prediction of mortality at a single-subject level with an accuracy of 72.72%.

Multiscale properties of instantaneous parasympathetic activity in severe congestive heart failure: A survivor vs non-survivor study

Barbieri, R.
2017-01-01

Abstract

Multifractal analysis of cardiovascular variability series is an effective tool for the characterization of pathological states associated with congestive heart failure (CHF). Consequently, variations of heartbeat scaling properties have been associated with the dynamical balancing of nonlinear sympathetic/vagal activity. Nevertheless, whether vagal dynamics has multifractal properties yet alone is currently unknown. In this study, we answer this question by conducting multifractal analysis through wavelet leader-based multiscale representations of instantaneous series of vagal activity as estimated from inhomogeneous point process models. Experimental tests were performed on data gathered from 57 CHF patients, aiming to investigate the automatic recognition accuracy in predicting survivor and non-survivor patients after a 4 years follow up. Results clearly indicate that, on both CHF groups, the instantaneous vagal activity displays power-law scaling for a large range of scales, from ≃ 0.5s to ≃ 100s. Using standard SVM algorithms, this information also allows for a prediction of mortality at a single-subject level with an accuracy of 72.72%.
2017
Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
9781509028092
Signal Processing; Biomedical Engineering; 1707; Health Informatics
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11311/1045369
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 1
social impact