Stability of ADC measurements in diffusion MRI: A multicenter phantom study using multi-point b-values.

Introduction: The apparent diffusion coefficient (ADC) from diffusion MRI provides critical insights into tissue microstructure and is influenced by various b-values. This study aimed to evaluate the impact of multiple b-values on ADC measurements using a multicenter observational approach with different MRI scanners and phantoms.

Methods: The study utilized a liquid isotropic phantom across eight different 1.5 T MRI scanners from five centers. The phantom contained four fluids with different viscosity and relaxation properties. ADC measurements were obtained at b-values of 0, 50, 500, 1000, and 1500 s/mm using standardized protocols, which involved applying consistent imaging parameters across all MRI scanners. Each scan was repeated three times, and statistical analyses, including Kruskal-Wallis and Bonferroni-corrected Mann-Whitney U tests, were performed to evaluate variability.

Results: Significant variations in ADC values were observed among different MRI scanners. Higher b-values generally resulted in lower ADC readings, reflecting restricted diffusion. The comparison between 2-point and multiple-point b-value techniques revealed that the latter provides more consistent ADC measurements. Bland-Altman analysis indicated a systematic difference between single-shot echo-planar imaging (ssEPI) and turbo spin echo (TSE) sequences, with TSE showing more homogeneous ADC values.

Conclusions: This study highlights the significant variability in ADC measurements across different MRI scanners and sequences, emphasizing the critical role of standardization in diffusion MRI protocols. The findings demonstrate that multiple-point b-value techniques offer greater consistency compared to 2-point methods, and TSE sequences yield more homogeneous ADC values than ssEPI.

Implication: These findings underscore the need for standardized protocols and multicenter guidelines to enhance the reliability of ADC as a diagnostic biomarker in clinical and research settings.