Work Package Leader: John Thornton & Amy McDowell (UCL)

The aim of this WP is to conduct a multi-site study across at least 5 institutions, applying the techniques and phantoms designed in WP1 and WP2 to the measurement of clinical MRI system/scanner performance. The study will evaluate the reproducibility of qMRI measurements using the project-developed novel qMRI phantoms and associated image acquisition and analysis procedures, to demonstrate the capability to calibrate and harmonise measurand maps from MRI scanners at different locations with different specifications. In addition, the study will quantify the limits of harmonisation between scanners at different locations as well as identifying and ranking the main drivers of inconsistency

This WP will focus on:

• Reviewing theoretical considerations, relevant information from the literature, and current protocols for qMRI in use at UCL, UHBW, and BHSCT, to identify the optimal imaging measurement strategies for each MRI measurand, to ensure the scope of the multi-site study is relevant to the user community and clearly defined.

• Developing standardised and optimised test protocols for accuracy and precision to evaluate the performance of commercially available MRI hardware. Care will be taken to ensure image data are produced in scanning times practically available at clinical sites (typically < 4 hours per session), based on several individual MRI measurements, and compatible with routinely available software. To support practical quality control measurements on hospital scanners where such long scanning sessions may not be routinely available, each measurand (e.g., T1, T2, proton density fat fraction, iron content and diffusivity etc.) will also be measured using a more rapid sequence closer in measurement time to methods used in routine clinical practice.

• Collecting benchmarking performance data (accuracy, precision, linearity, reproducibility) from
intensive longitudinal measurements performed using reference clinical MRI scanner systems at
project partner sites and the test protocols.

• Collecting benchmarking performance data (accuracy, precision, linearity, reproducibility) to generate baseline and more limited longitudinal measurements across a wide range of European installed commercial clinical 1.5 T and 3 T MRI machines.

• Investigating, designing, and providing preliminary validation of a suite of practical test protocols suitable for bridging the gap between measuring calibrated reference media under optimal reference MRI conditions, to clinical in vivo measurements in human tissue in clinical applications. This will provide guidelines for obtaining clinical measurements of commonly encountered MRI measurands (T1, T2, proton density fat fraction, iron content and diffusion) with optimal precision and accuracy in vivo.