Direct Cardiac T1 Mapping with Subspace Modeling and Free-breathing Data Acquisition

Area of Science:

• Cardiovascular Imaging
• Medical Physics
• Biomedical Engineering

Background:

• Cardiac T1 mapping is crucial for assessing cardiomyopathies using MRI.
• Conventional MRI T1 mapping requires breath-holding, posing challenges for some patients.
• Developing free-breathing techniques is essential for broader clinical applicability.

Purpose of the Study:

• To present a novel T1 estimation framework for cardiac T1 mapping using free-breathing data acquisition.
• To enable direct T1 estimation from undersampled k,t-space data.
• To improve the accuracy and feasibility of cardiac T1 mapping.

Main Methods:

• Utilized a free-breathing, ECG-gated, inversion-recovery fast low-angle shot (FLASH) sequence for sparse (k,t)-space data acquisition.
• Incorporated a T1 relaxation model into a direct reconstruction framework for end-to-end T1 mapping.
• Employed the alternating direction method of multipliers (ADMM) algorithm to solve the image reconstruction problem, including subproblems for low-rank constraint, parametric fitting, and TV-based denoising.

Main Results:

• The proposed direct T1 estimation method demonstrated reduced bias and variance compared to indirect approaches in simulations and in-vivo experiments.
• Numerical simulations and in-vivo experiments validated the performance of the direct approach.
• The direct method showed significant benefits in T1 estimation accuracy.

Conclusions:

• Direct cardiac T1 mapping with subspace modeling enables accurate free-breathing acquisition.
• The framework integrates MR physics knowledge and subspace modeling for enhanced cardiac T1 mapping.
• This approach holds potential for improved quantitative assessment of cardiomyopathies.