Volume 54, Issue 3 pp. 350-357
Original Article

Mathematical detection of aortic valve opening (B point) in impedance cardiography: A comparison of three popular algorithms

Javier Rodríguez Árbol

Corresponding Author

Javier Rodríguez Árbol

Brain, Mind, and Behavior Research Center, University of Granada, Granada, Spain

Address correspondence to: Mr. Javier Rodríguez Árbol, Centro de Investigación Mente, Cerebro y Comportamiento, Campus Universitario de Cartuja s/n, 18071 Granada, Spain. E-mail: [email protected]Search for more papers by this author
Pandelis Perakakis

Pandelis Perakakis

Brain, Mind, and Behavior Research Center, University of Granada, Granada, Spain

Search for more papers by this author
Alba Garrido

Alba Garrido

Brain, Mind, and Behavior Research Center, University of Granada, Granada, Spain

Search for more papers by this author
José Luis Mata

José Luis Mata

Brain, Mind, and Behavior Research Center, University of Granada, Granada, Spain

Search for more papers by this author
M. Carmen Fernández-Santaella

M. Carmen Fernández-Santaella

Brain, Mind, and Behavior Research Center, University of Granada, Granada, Spain

Search for more papers by this author
Jaime Vila

Jaime Vila

Brain, Mind, and Behavior Research Center, University of Granada, Granada, Spain

Search for more papers by this author
First published: 03 December 2016
Citations: 35

This research was funded by grants to JRÁ (FPU-MED AP-2010-4126), MCF-S (PSI2013-43777-P), and JV (PSI2011-28530) from the Spanish Ministry of Education and Culture and Spanish Ministry of Economy and Competitiveness. We are grateful to Eva-Maria Loch for her help in improving the writing style of the manuscript.

Abstract

The preejection period (PEP) is an index of left ventricle contractility widely used in psychophysiological research. Its computation requires detecting the moment when the aortic valve opens, which coincides with the B point in the first derivative of impedance cardiogram (ICG). Although this operation has been traditionally made via visual inspection, several algorithms based on derivative calculations have been developed to enable an automatic performance of the task. However, despite their popularity, data about their empirical validation are not always available. The present study analyzes the performance in the estimation of the aortic valve opening of three popular algorithms, by comparing their performance with the visual detection of the B point made by two independent scorers. Algorithm 1 is based on the first derivative of the ICG, Algorithm 2 on the second derivative, and Algorithm 3 on the third derivative. Algorithm 3 showed the highest accuracy rate (78.77%), followed by Algorithm 1 (24.57%) and Algorithm 2 (13.82%). In the automatic computation of PEP, Algorithm 2 resulted in significantly more missed cycles (48.57%) than Algorithm 1 (6.3%) and Algorithm 3 (3.5%). Algorithm 2 also estimated a significantly lower average PEP (70 ms), compared with the values obtained by Algorithm 1 (119 ms) and Algorithm 3 (113 ms). Our findings indicate that the algorithm based on the third derivative of the ICG performs significantly better. Nevertheless, a visual inspection of the signal proves indispensable, and this article provides a novel visual guide to facilitate the manual detection of the B point.

The full text of this article hosted at iucr.org is unavailable due to technical difficulties.