Motion-robust high-resolution 3D-GRASE pseudo-continuous ASL perfusion imaging at 3T

Purpose: High resolution arterial spin labeling (ASL)-based perfusion imaging is desirable for many applications but suffers from low SNR and motion sensitivity due to the necessity for highly segmented readouts. Acquisition and reconstruction strategies were optimized for motion-robust high-resolution pseudo-continuous arterial spin labeling (pCASL) of the brain at 3T using segmented 3D Gradient and Spin Echo (3D-GRASE) at 2.0-mm isotropic resolution. Methods: A high-resolution 3D-GRASE pCASL sequence was implemented at 3T and evaluated with a focus on the interaction between background suppression strength and motion-corrected reconstruction, together with the fat suppression and control-tag subtraction strategies that strong suppression requires. Motion correction was performed using a novel motion-compensated (alignedSENSE-based) reconstruction that simultaneously estimates the motion and image, accounting for contrast changes between tag and control data. Eight healthy volunteers were scanned under motion-free and cued motion conditions. Image quality was assessed against motion-free reference images using peak signal-to-noise ratio (PSNR) and a structural similarity index measure (SSIM). Results: The optimized acquisition-reconstruction pipeline enabled robust 2.0-mm isotropic pCASL at 3T and showed improved resistance to motion-related degradation. Strong BGS provided near-complete suppression of static tissue, and water excitation reduced lipid contamination and improved coil sensitivity estimation. The regularized alignedSENSE motion correction further improved perfusion-image quality, increasing SSIM by 18.7% under Strong BGS. Conclusion: Robust high-resolution pCASL at 3T can be achieved through combined optimization of acquisition and reconstruction. The combination of Strong BGS, water excitation, reference-superimposed processing, and regularized motion correction mitigates motion-related degradation and yields higher-quality perfusion images for high-resolution ASL.