HAITCH: A framework for distortion and motion correction in fetal multi-shell diffusion-weighted MRI.

Area of Science:

• Neuroimaging
• Biomedical Engineering
• Developmental Neuroscience

Background:

• Fetal motion and magnetic field inhomogeneities introduce artifacts in diffusion MRI (dMRI).
• These artifacts compromise data quality and reliability in high-angular resolution fetal dMRI.
• Existing tools cannot effectively process multi-shell high-angular resolution fetal dMRI data.

Purpose of the Study:

• To present High Angular resolution diffusion Imaging reconsTruction and Correction approacH (HAITCH), a tool for correcting and reconstructing fetal dMRI data.
• To address the limitations of existing methods in handling motion-induced artifacts in fetal dMRI.
• To improve the fidelity and reliability of fetal dMRI analysis for microstructure and tractography.

Main Methods:

• Developed HAITCH, an open-source framework for fetal dMRI data processing.
• Incorporated blip-reversed dual-echo acquisition for dynamic distortion correction.
• Implemented advanced model-free motion correction and outlier detection for robust reconstruction.
• Optimized multi-shell design for enhanced information capture and motion tolerance.

Main Results:

• HAITCH successfully corrects artifacts and reconstructs high-fidelity fetal dMRI data.
• Demonstrated significant improvements in data quality across diverse fetal ages and motion levels.
• Validated HAITCH's effectiveness on real fetal dMRI scans, outperforming existing tools.

Conclusions:

• HAITCH is the first publicly available tool for correcting and reconstructing multi-shell high-angular resolution fetal dMRI data.
• The framework enables reliable advanced diffusion modeling, including fiber orientation distribution function estimation.
• HAITCH facilitates more accurate analysis of fetal brain microstructure and tractography under challenging conditions.