Multiresolution imaging using golden angle stack-of-stars and compressed sensing for dynamic MR urography

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Renal Physiology
• Medical Diagnostics

Background:

• Accurate estimation of Glomerular Filtration Rate (GFR) is crucial for diagnosing and managing kidney disease.
• Current methods for GFR estimation have limitations.
• Novel imaging techniques are needed for non-invasive GFR assessment.

Purpose of the Study:

• To develop and validate a novel multiresolution MRI methodology for accurate in vivo GFR estimation.
• To assess the clinical utility of the proposed MRI technique.

Main Methods:

• Utilized a 3D golden-angle radial stack-of-stars (SoS) MRI trajectory for data acquisition.
• Employed multiresolution reconstruction with 1-s temporal resolution for arterial input function (AIF) and 4-s temporal resolution for renal dynamic data using compressed sensing (CS).
• Validated the method with simulations and compared MRI-based GFR estimates to serum creatinine-based eGFR in 10 subjects.

Main Results:

• The 4-s CS images reduced artifacts and noise; the 1-s AIF minimized GFR estimation errors.
• No statistically significant difference was found between MRI-based total GFR and serum creatinine-based eGFR (P=0.92).

Conclusions:

• Demonstrated the feasibility of multiresolution MRI with a golden angle radial SoS scheme for accurate GFR estimation.
• The technique produces diagnostic quality dynamic renal images in vivo.
• This method offers a promising non-invasive approach for GFR assessment.