About this Research Topic
Despite the initial promise of parametric native T1 and T2 mapping to provide a quantitative pixel-wise myocardial tissue phenotyping without the need for GBCA, the lack of standardization of mapping sequences and thresholds for health and disease has limited its translation into routine clinical use. Instead of relying on specific myocardial thresholds, it is now possible to automatically identify myocardial disease patterns by artificial intelligence (AI) approaches that exploit the existing myocardial signal features associated with fibrosis and oedema in non-contrast steady-state free precession cine imaging and T1 and T2 mapping.
The research community has been optimistic regarding the impact that these AI-based solutions for automatic image analysis can have on preventing unnecessary gadolinium exposition, shortening scan time, and reducing operator dependency thereby facilitating the widespread use of CMR namely in low-income countries. In the field of CMR, how can we ensure that clinical decisions (e.g., undergoing invasive coronary angiography or implanting a cardioverter defibrillator) based on these models are done the right way and for the right reasons?
This research topic was designed to provide an opportunity for the authors to contribute with gadolinium-free AI strategies to phenotype the myocardium. The topic will be kept broad, however, studies with an emphasis on the interpretability vs explainability of the proposed AI imaging models are welcome.
Keywords: MRI, gadolinium, AI, tissue characterization, magnetic resonance imaging
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.