ANR 2025
Project coordination
Pr olivier Baud, Chef de Service, Service de Médecine et Réanimation Néonatales de Port-Royal, Hôpital Universitaire Cochin, DMU FeMEO, AP-HP. Centre.
Partnership
- CRESS INSTITUT NATIONAL DE LA SANTE ET DE LA RECHERCHE MEDICALE
- CREATIS INSTITUT NATIONAL DES SCIENCES APPLIQUÉES LYON
- DRCI Direction de la Recherche et de l’Innovation de l’ AP-HP
Abstract
White matter lesions are one of the main risk factors for neurodevelopmental disorders (NDDs) in premature babies, and among these, punctate white matter lesions (PWMLs) are the most frequent. MRI is the reference brain imaging technique, but it suffers from limitations (poor accessibility, low portability), which reduce its contribution to early systematic screening for NDDs in all premature babies at risk. In contrast, conventional transfontanellar ultrasound (US) enables prospective detection of brain lesions, but only detects the most severe lesions, and is not efficient for the detection or volumetric quantification of cerebral anatomical structures. However, recent progress has been made with 3D US, paving the way for a new routine screening modality. The project aims to develop and test US quantification tools in a population of very premature infants, imaged from birth to term equivalent age, to develop algorithms for automatic detection of PWML and segmentation of certain structures with high predictive value (thalami, lateral ventricles). Using artificial intelligence, these algorithms will be integrated into pre-industrial research software with clinical implementation in the short term (WP2&3). Recruited children will be assessed using a developmental scale at the age of 2 to evaluate the contribution of longitudinal quantitative 3D US compared with MRI in predicting NDDs (WP4). The creation of this new imaging modality at bedside should have a major impact on optimizing the longitudinal detection of brain lesions and identifying biomarkers of NDDs associated with prematurity.