Practical signal-to-noise ratio quantification for sensitivity encoding: application to coronary MR angiography

Area of Science:

• Medical Imaging
• Cardiovascular Imaging
• Image Reconstruction

Background:

• Parallel imaging significantly accelerates magnetic resonance angiography (MRA) acquisition.
• Spatially varying noise introduced by parallel imaging reconstruction complicates accurate signal-to-noise ratio (SNR) quantification.
• Robust SNR assessment is crucial for evaluating image quality in coronary MRA.

Purpose of the Study:

• To develop and validate a practical method for quantifying SNR in coronary MRA acquired with parallel imaging.
• To assess the performance of the proposed SNR quantification method in phantom and in vivo studies.
• To evaluate the impact of parallel imaging on SNR in different tissues within coronary MRA.

Main Methods:

• A novel method was developed involving a fast noise scan with disabled radiofrequency pulses, cardiac triggering, and navigator gating.
• The method's accuracy was evaluated against a reference standard using multiple repetitions in phantom studies.
• In vivo SNR measurements were performed on human coronary MRA, comparing myocardium and posterior skeletal muscle.

Main Results:

• The proposed SNR quantification method showed less than 10.1% deviation from the reference method for geometry factors up to 2.
• The noise scan required only 30 seconds for a 10-minute coronary MRA acquisition.
• Myocardium exhibited higher signal but lower SNR compared to posterior skeletal muscle due to spatially varying noise.

Conclusions:

• SNR quantification using a fast noise scan is a validated and user-friendly approach for 3D coronary MRA with parallel imaging.
• The method is reliable when the geometry factor is low.
• This technique facilitates accurate image quality assessment in parallel imaging-based coronary MRA.