Enhanced Quantitative Susceptibility Mapping of Preterm Infant Brains via Nonlinear Total Field Inversion and Multi-parameter Regularization

Motivation: Preterm birth is a leading cause of neurodevelopmental disorders. Goal(s): Quantitative Susceptibility Mapping (QSM) is a valuable tool for studying brain development in preterm infants, but its reconstruction is susceptible to artifacts. Approach: To address these limitations, we propose a novel nonlinear preconditioned total field inversion (TFI) method, incorporating prior information from magnitude and R2* data. Results: Our method effectively mitigates noise and phase wrapping, leading to more accurate and robust QSM reconstructions. This enables the investigation of the spatiotemporal evolution of iron and myelin in specific brain regions of preterm infants, facilitating early identification of developmental abnormalities. Impact: Preliminary applications in preterm infants indicate that proposed approach can successfully model the essential spatiotemporal evolution of susceptibility. This advancement holds promise for establishing a framework for tracking brain maturation in preterm-populations, setting a new way in early neurodevelopmental assessment.