Quality assurance in quantitative MRI

Abstract

There is a continuously evolving paradigm shift in MRI, that of quantitative MRI (qMRI). Therefore, a gradual transition exists from a simplified, phenomenological and descriptive procedure to a quantitative measurement used for early diagnosis, disease evaluation and monitoring, as well as for therapeutic guidance. There is a plethora of parameters that can be used as quantitative markers in MRI, ranging from relaxation times to indices related to complex mechanisms such as diffusion and perfusion. The advent of qMRI has introduced new concepts in clinical MRI practice such as sensitivity, accuracy and specificity. The measurement precision and accuracy can be determined through a carefully designed program of quality control employing both phantoms and healthy volunteers. The need for high specificity lies in choosing the optimal parameter to be measured and monitored. The main requirement for increasing the measurement’s sensitivity and accuracy and for achieving high sensitivity is to fully understand the underlying q-MRI mechanism. With regard to the various sources of error and inaccuracy involved, q-MRI can be divided into two sub-processes: data collection and data analysis. A quality assurance program has to address issues related to both sub-processes. The second one refers to post-processing and, thus, it is a repeatable process which can adopt suitable measures for mitigating pertinent inaccuracies. The former refers to scanning. Spatial distortions and signal inaccuracies stem from many factors related to the measurement conditions. It is essential, therefore, to characterize, reduce and correct these distortions and inaccuracies, so as not to adversely affect quantitative results.