Abstract Background Apparent Diffusion Coefficient (ADC) as measured by diffusion weighted imaging is known to negatively correlate with prostate tumor aggressiveness. Heterogeneity in system and protocol performance causes potential variability in ADC acquired across a large scanner network, prompting a need to evaluate quantitative ADC from a prostate‐specific MR diffusion protocol as part of quality assurance (QA). Due to the temperature dependence of ADC, repeatability and reproducibility assessments typically require phantoms to maintain a temperature of 0°C, imposing a considerable burden when assessing large numbers of scanners. Purpose To develop a QA procedure at room temperature for assessing the reproducibility of ADC measured by our prostate diffusion MRI protocols, by employing a model relating ADC and phantom gel concentration to temperature. Methods A diffusion phantom was imaged with our clinical prostate diffusion protocols on 1.5 T ( n = 4) and 3 T ( n = 4) scanners in separate geographical locations ( n = 4). Phantom temperature was not strictly controlled but measured immediately before and after each acquisition. Regions of interest were drawn in phantom compartments to produce ADC measurements at varying phantom gel concentrations. The ADC temperature model was applied using measured temperatures to produce normalized ADC measurements at a reference temperature of 20°C. Percent difference and coefficient of variation (CV) calculations were performed on the measurements with and without normalization. Results Temperature normalization reduced the range of ADC percent difference across scanners. The maximum CV of clinically relevant ADC values was reduced from 5.0% to 3.4% when normalized to the temperature model, while CV of pure water was reduced from 3.3% to 1.3%. Conclusion We have developed a QA procedure that incorporates a temperature model to obviate the need for strict temperature control of a diffusion phantom, facilitating ADC reproducibility assessments of a prostate diffusion MRI protocol across a large scanner network.
Hwang et al. (Thu,) studied this question.