Loading...

Causal Connectivity Effects between Cardiovascular Signals

Bahrami Yarahmadi, Shaghayegh | 2022

126 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 55923 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Shamsollahi, Mohammad Bagher
  7. Abstract:
  8. Cardiovascular signals are used to check the performance of the cardiovascular system. Researchers have always been interested in processing cardiovascular signals to achieve goals like diagnosing arrhythmias and heart diseases. The presence of causal or non-causal relationships between these signals can be used to evaluate the cardiovascular system's performance. Cardiovascular signals, such as ECG signal, RR series and systolic arterial blood pressure, chest volume or respiratory force, and blood oxygen concentration signals are used to estimate cardiovascular connectivities. There are different approaches to assessing causal relationships; among these approaches, methods based on Granger causality, whose basic formulation is based on the MVAR model, have been of great interest. On the other hand, the linear assumption taken in the original formulation of Granger causality is not necessarily the correct assumption and nonlinear extensions of Granger causality are more realistic. In the present study, the causal relationships between cardiovascular signals have been estimated by criteria defined based on a nonlinear extension of Granger causality. The extension above uses NARX modeling to estimate and calculate the criteria of causal interactions. The investigation of the causal relationships between cardiovascular signals under different body conditions has been realized and the results have shown that causal interactions are not the same under different body states. Also, a nonlinear approach previously used in causal brain relationships was used to evaluate the cardiovascular and bidirectional causal effects were detected between each pair of investigated parameters. In the meantime, the dominant direction of the causal interaction between heart rate and respiration is from respiration to heart rate, and both parameters of heart rate and respiration affect the blood oxygen level. Still, respiration is a more determining factor in determining the oxygen level in the blood
  9. Keywords:
  10. Effective Connectivity ; Granger Causality ; Nonlinear Autoregressive with Exogenous Input Network (NARx) ; Cardiovascular Signals ; Parametric Causality ; Bivariate Causality ; Nonlinear Causality

 Digital Object List

 Bookmark

No TOC