Model-based acceleration of look-locker T1 mapping.

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Quantitative Imaging
• Biomedical Engineering

Background:

• Quantitative T1 mapping in MRI is crucial for comparing tissue properties across subjects and scanners.
• Conventional T1 mapping methods suffer from long scan times, limiting clinical adoption.
• Accelerated techniques are needed to make quantitative MRI more feasible in routine clinical practice.

Purpose of the Study:

• To develop and validate an accelerated Inversion-Recovery Look-Locker (IR-LL) T1 mapping technique.
• To enable fast and accurate T1 measurements with reduced acquisition time.
• To assess the accuracy, quality, and reproducibility of the proposed method.

Main Methods:

• A model-based algorithm was developed to reconstruct T1 maps from highly undersampled, radially acquired IR-LL data.
• The reconstruction enforced an exponential relaxation model after a single global inversion pulse.
• Validation was performed using a segmented IR experiment as the gold standard in phantoms and in vivo.

Main Results:

• A T1 map with 1.6mm in-plane resolution was reconstructed in just 6 seconds.
• Phantom validation showed T1 differences <2.6% compared to the gold standard.
• In vivo studies in healthy volunteers demonstrated T1 differences <5.5% in white and gray matter, with comparable literature values and high visual quality.

Conclusions:

• The presented model-based reconstruction enables highly accelerated, fast, and accurate T1 mapping using radial IR-LL acquisitions.
• This approach eliminates the need for time-consuming segmentation, offering a promising solution for clinical quantitative MRI.
• The method demonstrated excellent reproducibility and visual quality, comparable to fully sampled references.