Advanced Models for Monitoring Stress and Development Trajectories in Premature Infants

This thesis focuses on the design of various automatic signal processing algorithms to extract information from physiological signals of preterm infants. Overall, the aim was to improve the neurodevelopmental outcome of the neonate. More specifically, three main research objectives were carried out. The first objective was to describe the maturation of neonates during their stay in the neonatal intensive care unit. The second objective was to assess the stress and pain in premature infants and their impact on the development of neonates. The third objective was to predict developmental disabilities, such as autism. The first part of this thesis presents an extensive overview of various developmental models to describe the maturation of premature infants. Three main strategies were proposed. The first strategy proposed an investigation of EEG connectivity networks. A variety of functional and effective connectivity methods were combined with an assessment of graph properties of the resulting EEG network. A set of topological and spectral indices were used to predict the age of the infants and to demonstrate that the functional and effective connectivity decreased with development, characterized by a shift from a small-world network to a more random network. The second strategy focused on the multifractal properties of the EEG signal, which were investigated to detect quiet sleep in premature infants on the one hand, and to describe developmental changes of the EEG on the other hand. The fractal paradigm showed a decrease of EEG regularity with increasing age. The last strategy reported the trends of heart-rate variability in premature infants to describe the development of the autonomic nervous system. Similar to the EEG analysis, the autonomic parameters showed a decrease of fractal regularity and an increase in short- and long-term variability of the tachogram. Most importantly, a full description of the autonomic maturation in premature has to take into account the roles of heart-drops, known as bradycardias, and new metrics to define the sympathovagal development. The second part focuses on the quantification of perinatal stress in premature infants. Firstly, the effect of stress load was investigated during bradycardia and hypoxic events. We established that stress-load and early-life experiences can enhance desaturations, EEG regularity and brain-heart connectivity during bradycardias. Secondly, we processed physiological background activity of premature infants without any pain elicitation. This unobtrusive approach showed a moderate association between physiological features and perinatal stress. More specifically, stress seemed to induce a more dysmature EEG, characterized by more discontinuity and slow-wave activity, a more synchronized EEG network, and a stronger brain-heart coupling. In addition, the effect of stress on preterm development during their hospital stay was examined and the discontinuous EEG seemed to persist throughout the infants' maturation. The third part reports an application of early-life EEG analysis, in which we aimed to predict neurodevelopmental disabilities, such as autism spectrum disorder. Based on EEG dysmature traits in the first month of life, such as discontinuity and slow-wave persistence, we showed that different classification models were capable to predict the autism diagnosis at 24 months and other developmental abnormalities (such as a lower cognitive outcome). The quantitative analysis of EEG confirmed the central role of dysmaturity attributes (the lower EEG entropy, the higher EEG regularity and a greater brain network resilience) as the most predictive attributes of developmental abnormalities and disorders.

Jaar van publicatie:2020

Toegankelijkheid:Open