Conformance of a 3T radiotherapy MRI scanner to the QIBA Diffusion Profile.

Purpose: To assess the technical performance of the apparent diffusion coefficient (ADC) on a dedicated 3T radiotherapy scanner, using a standardized phantom and sequences. Investigations into factors that could impact the technical performance of ADC in the clinic were also completed, including changing the slice-encoded imaging direction and the reference sample ADC value.
Methods: ADC acquisitions were performed monthly on an isotropic diffusion phantom over 1 year. Measurements of ADC %bias, coefficients of variation for short-/long-term repeatability and precision (CV _ST /CV _LT and CV _P ), and b-value dependency (Dep _b ) were calculated. The measurements were then assessed according to the Quantitative Imaging Biomarker Alliance (QIBA) Diffusion Profile specifications.
Results: The average of all measurements over the year was within Profile recommended ranges. This included when testing was performed in different imaging directions, and on samples that had different ADC reference values (0.4-1.1 μm ² /ms). Results in the axial plane for the central water vial included a bias of +0.05%, CV _ST /CV _LT /CV _P  = 0.1%/ 0.9%/0.4% and Dep _b  = 0.4%.
Conclusions: The technical performance of ADC on a radiotherapy dedicated MRI scanner over the course of 12 months was considered conformant to the QIBA Profile. Quantifying these metrics and factors that may affect the performance is essential in progressing the use of ADC clinically: ensuring that the observed change of ADC in a tissue is due to a physiological response and not measurement variability.
(© 2022 The Authors. Medical Physics published by Wiley Periodicals LLC on behalf of American Association of Physicists in Medicine.)