Accelerated 3D-GRASE imaging improves quantitative multiple post labeling delay arterial spin labeling.

Area of Science:

• Magnetic Resonance Imaging
• Neuroimaging
• Medical Physics

Background:

• Quantitative arterial spin labeling (ASL) provides non-invasive measurement of cerebral blood flow.
• Accelerated imaging techniques are crucial for reducing scan times in ASL.
• 3D-GRASE sequences offer fast volumetric coverage but can be limited by acceleration artifacts.

Purpose of the Study:

• To evaluate the impact of accelerated single-shot 3D-GRASE acquisition on quantitative ASL.
• To assess the effects on perfusion-weighted signal-to-noise ratio per unit scan time (TSNR_PW) and quantification accuracy.
• To compare single and multiple post-labeling delay (PLD) acquisitions under accelerated conditions.

Main Methods:

• Five subjects underwent 3T MRI using pseudo-continuous ASL (PCASL) with accelerated 3D-GRASE.
• Simulated and implemented 3-inversion pulse background suppression.
• Compared TSNR_PW for single PLD using segmented, GRAPPA, and CAIPIRINHA sampling.
• Evaluated multiple PLD acquisitions (no acceleration, CAIPIRINHA, CAIPIRINHA with doubled PLDs) for cerebral blood flow and arterial transit time fit uncertainties.

Main Results:

• CAIPIRINHA sampling increased TSNR_PW by 11% compared to GRAPPA for single PLD.
• Non-accelerated scans showed 35% higher TSNR_PW than GRAPPA.
• CAIPIRINHA sampling with multiple PLDs (6/12) reduced CBF fit uncertainty by 14%/16% (gray matter) and 34%/36% (white matter) compared to non-accelerated scans.

Conclusions:

• Accelerated single-shot 3D-GRASE with PCASL achieves lower quantification uncertainties than time-matched segmented ASL.
• This accelerated approach enables state-of-the-art ASL methods within clinically feasible scan times.
• Acceptable image quality with minimal blurring and motion artifacts is achievable.