Related Experiment Video
Updated: Apr 22, 2026
High-resolution Functional Magnetic Resonance Imaging Methods for Human Midbrain
Published on: May 10, 2012
Ultrafast Infant Brain Quantitative MRI Using Overlapping-Echo Acquisition with Volumetric Physical Simulation of
Objective:
Quantitative MRI (qMRI) is sensitive to brain microstructural and metabolic changes; however, existing techniques often unsuitable for assessing postnatal brain development due to prolonged scan time, non-ideal imaging conditions, and severe infant motion. In this study, we propose a robust qMRI framework tailored for infant brain imaging to address abovementioned limitations.
Methods:
The proposed framework integrates multiple overlapping-echo detachment (MOLED) acquisition and synthetic data-driven deep learning quantitative reconstruction. To overcome non-ideal imaging conditions, a volumetric physical simulation (VoluSimu) method was proposed to generate physics-informed synthetic training data, ultimately enabling robust T2, T2 *, and T2' mapping of infant brain. Numerical and phantom experiments were used to validate the quantitative accuracy and robustness of proposed framework under various non-ideal imaging conditions. Additionally, we applied the framework to a cohort of 30 infants (0-1 year), obtaining their T2, T2 *, and T2' maps to quantitatively characterize early brain development.
Results:
Validation studies demonstrated that, despite imperfect RF pulses and B0 inhomogeneity, network trained on VoluSimu-generated synthetic data provided accurate quantitative results. In in vivo study, MOLED-derived maps were free of motion artifacts. T2 and T2 * values showed mono-exponential age-related changes, while T2' values decreased linearly in white matter and cortical gray matter.
Conclusion And Significance:
The proposed framework enables high-fidelity, ultrafast infant brain qMRI, overcoming challenges posed by non-ideal imaging conditions and motion. The resulting quantitative parameters offer valuable insights into microstructural and metabolic changes associated with brain maturation, demonstrating the potential of the proposed framework for longitudinal monitoring of early brain development.
