Comparison of image quality evaluation methods for magnetic resonance imaging using compressed sensing-sensitivity...

Area of Science:

• Medical Imaging
• Magnetic Resonance Imaging (MRI)

Background:

• Compressed SENSE (CS-SENSE) is an accelerated MRI technique.
• Assessing image quality in accelerated MRI is crucial for clinical application.
• Quantitative metrics and visual evaluation scores are used to evaluate MRI quality.

Purpose of the Study:

• To compare quantitative image quality values with visual evaluation scores for CS-SENSE MRI.
• To determine the effectiveness of different quantitative metrics in assessing image quality for T1-weighted and T2-weighted images acquired with CS-SENSE.
• To clarify the differences in evaluating image quality between T1-weighted and T2-weighted images using CS-SENSE.

Main Methods:

• Acquired T1-weighted images (T1WI) and T2-weighted images (T2WI) using a 3D-printed phantom and the CS-SENSE method.
• Calculated quantitative values: signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), structural similarity (SSIM), and scale-invariant feature transform (SIFT).
• Calculated visual evaluation score (VES) and determined correlation coefficients between quantitative values and VES.

Main Results:

• Established correlations between quantitative metrics (SNR, CNR, SSIM, SIFT) and visual evaluation scores (VES) for CS-SENSE images.
• Demonstrated that CNR is effective for evaluating T1-weighted image quality.
• Identified SNR, CNR, and SIFT as suitable metrics for assessing T2-weighted image quality variations.

Conclusions:

• Image quality variations in CS-SENSE T1WI and T2WI can be quantitatively assessed.
• Specific quantitative metrics show varying degrees of effectiveness for different image contrasts (T1WI vs. T2WI).
• This study provides insights into optimizing image quality assessment for accelerated MRI techniques.