Reproducibility of Ultrafast High-Resolution MRSI at 3T Using SPICE

Motivation: Understanding the intra-session and inter-session reproducibility is essential for practical applications of MRSI, especially in longitudinal studies. Goal(s): To evaluate the intra-session and inter-session reproducibility of SPICE for ultrafast high-resolution MRSI. Approach: We recruited 10 subjects from 2 sites in a two-session two-repetition reproducibility study, with between-session interval 12 to 22 days. MRSI results were analyzed in 33 brain regions for both intra-session and inter-session reproducibility. Results: Low intra-session and inter-session coefficients of variations (CVs) were obtained, with intra-session CVs slightly lower. ANOVA results indicated significant effects of the subject factor, but not the session and repetition factors. Impact: This study demonstrated very good intra-session and inter-session reproducibility of SPICE. The results indicate that changes larger than the reported intra-session CVs (for cross-sectional studies) or inter-session CVs (for longitudinal studies) are likely attributable to biological metabolic alterations.