Title: Multivariate ECG analysis for apnoea–bradycardia detection and characterisation in preterm infants

Authors: Miguel Altuve, Guy Carrault, Julio Cruz, Alain Beuchee, Patrick Pladys, Alfredo I. Hernandez

Addresses: Department of Industrial Technology, Simon Bolivar University, Caracas 1080, Venezuela; Universite de Rennes 1, LTSI, Rennes, F-35000, France; INSERM, U642, Rennes, F-35000, France. ' Universite de Rennes 1, LTSI, Rennes F-35000, France; and INSERM, U642, Rennes F-35000, France. ' Department of Electronic and Circuit, Simon Bolivar University, Caracas 1080, Venezuela. ' Universite de Rennes 1, LTSI, Rennes, F-35000, France; INSERM, U642, Rennes, F-35000, France; Pole Medico-Chirurgical de Pediatrie et de Genetique Clinique, Neonatologie, Rennes F-35000, France. ' Universite de Rennes 1, LTSI, Rennes, F-35000, France; INSERM, U642, Rennes, F-35000, France; Pole Medico-Chirurgical de Pediatrie et de Genetique Clinique, Neonatologie, Rennes F-35000, France. ' Universite de Rennes 1, LTSI, Rennes F-35000, France; INSERM, U642, Rennes F-35000, France

Abstract: An analysis of the information content of new features derived from the electrocardiogram of preterm infants is presented. Beat detection and segmentation methods, specifically adapted to this population through evolutionary algorithms, led to an improved sensitivity and positive predictive value. RR interval, R-wave amplitude and QRS duration time-series were extracted at rest (T1), before (T2), during (T3) and after (T4) apnoea-bradycardia episodes. Significant variations for T1-T2 vs. T1-T3 and T1-T3 vs. T1-T4 were observed. Results reveal changes in the R-wave amplitude and QRS duration at the onset and termination of apnoea-bradycardia episodes which could be useful for their detection and characterisation.

Keywords: apnoea; bradycardia; preterm infants; electrocardiograms; ECG analysis; beat detection; ECG segmentation; wavelet transforms; evolutionary algorithms.

DOI: 10.1504/IJBET.2011.039201

International Journal of Biomedical Engineering and Technology, 2011 Vol.5 No.2/3, pp.247 - 265

Published online: 21 Jan 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article