Detecting microstructures in magnetic resonance imaging with a spin-lock pulse sequence

Abstract

Elevated micrometer-scale iron deposits in the brain are a crucial early detection marker for numerous neurodegenerative diseases. Although iron deposits exist below conventional magnetic resonance imaging resolution, sub-voxel information on their spatial properties can be encoded into the MR signal through magnetic susceptibility differences and diffusion effects. Spin-lock pulse sequences have recently emerged as a powerful tool sensitive to diffusion-mediated dephasing, characterized by the time constant T_1ρ. By employing continuous low-frequency radiofrequency pulses, signal dynamics can be sensitized to motions in the order of sub-kilohertz, rendering it sensitive to the effect of diffusion. In this work, the potential of microstructure characterization with T_1ρ was explored through simulation and phantom experiments. A Monte Carlo simulation of a conventional spin-lock pulse showed high sensitivity to microbead radius, concentration, and susceptibility shift through R_1ρ dispersion magnitude and inflection point. Phantom experiments of a balanced and refocused spin-lock pulse demonstrated minimal changes in relaxation rate, suggesting that a considerable susceptibility gradient must be present before signal dynamics are affected. By overcoming current experimental limitations, spin-lock pulse sequences hold great promise as reliable tools for probing structures of micrometer size.