A Low-Field MRI Dataset For Spatiotemporal Analysis of Developing Brain

Area of Science:

• Neuroimaging
• Developmental Neuroscience
• Medical Imaging

Background:

• High-field MR imaging for infant brain studies faces accessibility and reproducibility challenges.
• Low-field MR imaging presents a safer, more portable, and cost-effective alternative for examining the developing brain.
• Understanding infant brain development is crucial for linking brain structure to behavioral changes.

Purpose of the Study:

• To demonstrate the feasibility of using low-field MR imaging for examining structural brain changes in infants.
• To present a novel dataset of low-field structural MR images of the infant brain.
• To address the scarcity of large, extended-span infant brain datasets for developmental trajectory analysis.

Main Methods:

• Acquisition of 100 T2-weighted structural MR images from infants using low-field MR.
• In-plane resolution of ~0.85 mm and slice thickness of ~6 mm.
• Atlas-based whole brain segmentation and volumetric quantification for developmental analysis.

Main Results:

• The presented low-field infant MR dataset enables the examination of brain structural changes in early postnatal life.
• Atlas-based analysis demonstrated the utility of the dataset for quantifying brain development features within the first 10 weeks.
• The dataset supports the development of routine low-field MR imaging pipelines for infant studies.

Conclusions:

• Low-field MR imaging is a feasible and advantageous tool for infant brain research.
• The created dataset facilitates the tracking of infant brain development trajectories.
• This work contributes to advancing accessible and reproducible neuroimaging methods for early life brain studies.