Description du sujet de thèse

Preterm birth affects approximately one in ten children globally and presents significant developmental risks. This PhD project proposes an interdisciplinary approach, combining neuroscience, developmental psychology, and advanced data science methods, to study premature children's developmental trajectories. The project has three main objectives: (1) identify distinct multilevel profiles in premature children aged 9–10 using genetic, brain, cognitive, and behavioral data through sophisticated machine learning techniques (including recurrent neural networks and temporal convolutional networks); (2) analyze their developmental trajectories to age 15 using latent class mixed models (LCMM); and (3) determine predictive factors through multivariate statistical analysis. Drawing on the ABCD® study cohort of over 10,000 children (including 2,000 premature), this research will leverage state-of-the-art statistical methods and machine learning models to deepen our understanding of preterm-born children's development. Clinically, identifying risk and protective factors through these advanced analytical approaches will guide the development of personalized early interventions. The innovative data science methodology developed during this PhD could also be adapted to various clinical applications.

Contexte de travail

The PhD candidate will be funded by the CNRS through the MITI (Mission pour les Initiatives Transverses et Interdisciplinaires) for 3 years to complete the project. The candidate will be based at the LaPsyDÉ Lab (https://www.lapsyde.com/, CNRS UMR 8240, Université Paris Cité) at La Sorbonne in Paris, France and will collaborate with Edouard Duchesnay, GAIA Brain Imaging and Data Science Lab (https://gaia.neurospin.fr/, CNRS UMR 9027, CEA) at Neurospin center in Saclay.

Contraintes et risques

No particular constraints or risks.