Assessing Nonlinear Properties in Breathing Signals from Preterm Infants

Xavier Navarro (1) , Fabienne Porée (1) , Alain Beuchée (1) , Guy Carrault (1)

1 LTSI - Laboratoire Traitement du Signal et de l'Image

Breathing signals from preterm infants are studied with the noise titration technique and surrogate-data analysis using nonlinear indexes from correlation dimension and sample entropy. We verify that most analyzed breathing time-series contains nonlinear dynamics, and their strengths become more significant as maturational aspects (weight and post-conceptional age) increase. This behavior is partially followed in variability-series extracted from five-minute breathing signals. We also observe that entropy and titration based tests were more robust than correlation dimension in short and noisy signals.

Mots clés: Preterm infants; Breathing; Nonlinear; Surrogate-data test; Noise titration

Domaines: Pédiatrie; Ingénierie biomédicale; Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Xavier Navarro, Fabienne Porée, Alain Beuchée, Guy Carrault. Assessing Nonlinear Properties in Breathing Signals from Preterm Infants. International Journal of bioelectromagnetism, 2013, 15 (1), pp.102 - 108. ⟨hal-00880239⟩