Accelerated T1ρ acquisition for knee cartilage quantification using compressed sensing and data-driven parallel...

Area of Science:

• Magnetic Resonance Imaging
• Biomedical Engineering
• Radiology

Background:

• Quantitative T1ρ imaging aids in early osteoarthritis detection.
• Clinical adoption is limited by lengthy scan times.
• Accelerated techniques are needed to improve efficiency.

Purpose of the Study:

• To evaluate accelerated T1ρ mapping for knee cartilage quantification.
• To assess the feasibility of combining compressed sensing (CS) and autocalibrating reconstruction for Cartesian sampling (ARC).
• To improve the clinical accessibility of quantitative T1ρ imaging.

Main Methods:

• A sequential combination of ARC and CS was employed for accelerated T1ρ map acquisition.
• Imaging was performed on phantom, ex vivo (porcine knee), and in vivo (human knee) samples using a GE 3T MR750 scanner.
• CS-accelerated T1ρ quantification was compared against non-accelerated methods.

Main Results:

• Accelerated acquisition using CS did not significantly alter quantification accuracy.
• The coefficient of variation for root mean squared error remained below 5% for in vivo measurements up to a net acceleration factor of 2.
• The accelerated method achieved a 25% faster acquisition time compared to the reference (ARC only).

Conclusions:

• This study represents the first in vivo application of CS for T1ρ quantification.
• The findings suggest that accelerated T1ρ mapping is a promising technique for clinical use.
• The developed method enhances the accessibility of quantitative imaging for osteoarthritis assessment.