par Pelc, Karine 
Président du jury Casimir, Georges
Promoteur Verhoeven, Caroline
Co-Promoteur Dan, Bernard
Publication Non publié, 2025-09-30

Président du jury Casimir, Georges

Promoteur Verhoeven, Caroline

Co-Promoteur Dan, Bernard

Publication Non publié, 2025-09-30
Thèse de doctorat
| Résumé : | Brain maturation is a lifelong process, but it is most intense in early life. It involves neurophysiological changes that relate to the establishment of neuronal networks, connectivity within these networks, and neural complexity. Gradually, brain functioning integrates environmental influences and becomes better organised and adapted to developmental demands. Preterm infants, especially those born extremely or very preterm, face an increased risk of adverse neurodevelopmental outcomes due to interference with early brain maturation. However, many extremely or very preterm infants show resilience and achieve typical developmental milestones. This population of extremely or very preterm infants without neurodevelopmental disorders has been underexplored.In this work, we explored selected aspects of brain maturation and the evolution of brain complexity in a large cohort of such infants in order to provide novel insights into developmental trajectories. Though technically difficult, high-density EEG is important in infants owing to highly localised cortical activity. Longitudinal recording is also particularly relevant to the study of maturational processes. We used a prospective exploratory longitudinal design, recording high-density EEG at six time-points from term-equivalent age to 24 months corrected age. We analysed the evolution of connectivity and complexity of spontaneous EEG activity through multiscale entropy. We then analysed auditory processing through event-related potentials in response to syllable sounds stimuli, including auditory discrimination using a pseudorandom oddball paradigm.Our hypotheses were as follows:1.The correlation of multiscale entropy in EEG signals increases with age.2.EEG complexity shows regional differentiation during the first 24 months.3.Auditory processing becomes localised within the auditory cortex by 6 months.4.Auditory discrimination shows marked refinement by 6 months.5.Extrauterine life impacts brain maturation at least up to term age.We found that the correlation between neural complexity, as measured by multiscale entropy in EEG signals, increases with age across the cortex, reflecting the progressive maturation of functional brain connectivity. The increase in complexity followed a posterior-to-anterior gradient, consistent with other approaches of neurological development that suggest that sensory networks mature earlier than prefrontal systems. Additionally, we found a narrowing of the topography of the initial component of the event-related response (infant-P1) and its cortical generator towards the auditory cortex during the first few months of life suggesting that infant-P1 may reflect early cortical sound detection. Infant-N1, which emerges consistently around 6 months, may reflect cortical processing of sound content. Over the first two years, infant-N1 source generators gradually organise along the auditory dorsal stream and the mirror neuron system, potentially linking this activity to the development of speech processing. Our analysis of mismatch responses revealed significant differences in auditory discrimination capabilities starting at 6 months, indicating infants' ability to detect phonetic differences. Contrary to our last hypothesis, we didn’t see a correlation (or only a weak one) between the duration of extrauterine life and both neural complexity and auditory processing at term age, possibly due to uncontrolled factors.Our work provides insights into several important aspects of brain maturation. We highlight the six-month milestone as a pivotal phase in sensory auditory development. Our results could help refine neurophysiological biomarkers for early development in preterm populations without neurodevelopmental disorders. Specifically, the combination of complexity metrics such as multiscale entropy with cortical event-related potentials such as response to pre-speech stimuli, has the potential to enhance brain function assessment, risk stratification, early intervention, and neurodevelopmental monitoring. |



