Magnetic resonance imaging (MRI) can provide detailed information about how blood flows through tissues. This makes quantitative perfusion MRI a promising tool for studying diseases such as cancer, where changes in tissue perfusion can provide information about the tissue itself, how a tumour responds to treatment, or how a disease develops over time.
Two particularly promising techniques for measuring perfusion are intravoxel incoherent motion (IVIM) MRI and dynamic contrast-enhanced (DCE) MRI. Despite many promising research studies, however, these techniques are not yet routinely used as quantitative clinical biomarkers. One important reason is that measurements made at different hospitals or on different MRI scanners can produce different results, even when imaging the same tissue.
The ReCreaTe project aims to address this problem by developing a standardized, vendor-neutral approach to quantitative perfusion MRI.
We will investigate the entire imaging pipeline, from how MRI data are acquired and reconstructed to how the final quantitative measurements are calculated. We will develop open-source acquisition, reconstruction, and analysis tools that can be used across different MRI systems and sites. This will allow us to identify which parts of the imaging pipeline contribute most to differences between scanners and hospitals, and determine how these differences can be reduced.
To test the approach, we will use healthy volunteers who travel between imaging sites, allowing us to measure the same person on different MRI scanners under controlled conditions. We will also use a perfusion phantom to provide a stable reference for quality assurance and to systematically investigate how different acquisition and reconstruction settings affect the measurements.
By comparing standardized protocols with the protocols currently used at individual sites, we will quantify how much reproducibility can be improved and determine which technical choices matter most. The project will be developed together with the international perfusion MRI community, with the aim of translating the results into practical implementations that can ultimately be incorporated by MRI vendors.
Our goal is to make quantitative perfusion MRI reliable and reproducible across scanners and hospitals. This would help move perfusion MRI from promising single-centre research studies towards robust biomarkers that can be used in large clinical trials and, ultimately, to support personalized treatment decisions.