Vendor-agnostic 3D multiparametric relaxometry improves cross-platform reproducibility.

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Medical Physics
• Biomedical Engineering

Background:

• Multiparametric relaxometry provides valuable tissue characteristics.
• Current MRI techniques face challenges in data harmonization across different scanners and vendors.
• There is a need for a standardized, cross-platform approach to relaxometry.

Purpose of the Study:

• To implement and evaluate a simultaneous T1 and T2 mapping technique (3D-QALAS) using the open-source Pulseq platform.
• To assess the cross-platform, multiparametric relaxometry technique's performance across different vendors and sites.
• To address the unmet need for data harmonization in MRI relaxometry.

Main Methods:

• Implemented 3D-QALAS using the vendor-agnostic Pulseq platform for simultaneous T1 and T2 mapping.
• Tested the technique on four 3T scanners from two vendors across two sites.
• Evaluated cross-scanner, cross-software version, cross-site, and cross-vendor variability using a phantom and human subjects.

Main Results:

• Pulseq-QALAS showed high linearity and correlation with reference values in a phantom (R² > 0.99).
• The Pulseq implementation significantly improved phantom T2 reproducibility (CV, 2.3% vs. 17%) compared to vendor-native sequences.
• Reduced cross-vendor variability in vivo, particularly for gray matter T2 values (CV, 2.3% vs. 5.9%).

Conclusions:

• An identical implementation of relaxometry techniques across platforms enhances measurement reproducibility.
• The Pulseq-based 3D-QALAS technique facilitates data harmonization for multiparametric MRI.
• Standardized approaches are crucial for reliable and comparable relaxometry data acquisition.