Optimizing MPPCA for Complex Sodium MRI: Component Analysis and Noise Characterization at 3T

Motivation: Complex sodium MRI provides unique metabolic insights but suffers from inherently low SNR (signal-to-noise ratio). While MPPCA denoising has proven effective for magnitude images, its application to complex sodium MRI data, particularly preserving phase information, remains unexplored. Goal(s): To optimize MPPCA denoising for complex sodium MRI by: (1) determining optimal component selection, (2) characterizing noise distributions, and (3) quantifying structural preservation. Approach: Applied MPPCA with varied components to phase-corrected complex sodium MRI data from 35 subjects. Analyzed denoising performance using PSNR, SSIM, and residual characteristics. Results: Achieved optimal denoising and residual analysis confirmed unbiased noise removal with approximately Gaussian distribution, maintaining anatomical integrity. Impact: This comprehensive validation establishes MPPCA as a robust denoising method for complex sodium MRI, providing specific guidelines for component selection and quality metrics. The improved image quality enhances sodium MRI's potential for clinical applications in metabolic and neurological assessments.