Accelerating T1ρ cartilage imaging using compressed sensing with iterative locally adapted support detection and...

Area of Science:

• Magnetic Resonance Imaging
• Biomedical Engineering
• Medical Physics

Background:

• Quantitative T1ρ (T1rho) MRI is crucial for cartilage evaluation.
• Current T1ρ quantification methods require long acquisition times.
• Accelerated imaging techniques are needed for clinical feasibility.

Purpose of the Study:

• To develop and validate an accelerated T1ρ quantification technique for cartilage.
• To combine compressed sensing (CS) with iterative locally adaptive support detection (LAISD) and JSENSE for faster T1ρ mapping.
• To assess the accuracy of accelerated T1ρ quantification compared to conventional methods.

Main Methods:

• Proposed a novel reconstruction approach combining LAISD and JSENSE for accelerated T1ρ imaging.
• Iterative reconstruction involved local support detection, CS reconstruction, and JSENSE sensitivity estimation.
• Evaluated the method using in vivo knee cartilage data from healthy volunteers at different T1ρ times (TSLs).

Main Results:

• Achieved T1ρ quantification from accelerated scans (acceleration factors 3 and 3.5) comparable to full scans.
• Demonstrated T1ρ errors below 1% across all cartilage compartments.
• Results showed T1ρ errors well within the reported in vivo reproducibility limits.

Conclusions:

• The proposed method significantly accelerates T1ρ quantification in cartilage MRI.
• Accuracy is maintained, enabling faster and more efficient cartilage assessment.
• This technique holds promise for facilitating clinical translation of quantitative cartilage MRI.