• cardiac magnetic resonance imaging
• image processing
• radiomics
• repeatability
• t1 mapping
• t2 mapping

Radiomics is a novel tool that involves the extraction of a large number of morphological and textural quantitative characteristics (i.e., radiomic features) from digital medical images. This quantitative information could then be used for diagnostic, prognostic or predictive purposes exploiting statistical or machine learning methods.
Recently, in cardiac magnetic resonance (CMR) imaging, T1 and T2 mapping techniques have enabled a quantitative assessment of myocardial tissue characteristics. Specifically, these techniques can be employed to evaluate myocardial diseases that alter the composition of myocardial tissue, and possibly its T1 and T2 relaxation times. Hence, T1 and T2 maps might be particularly suitable for radiomic analysis.
Despite the increasing interest in radiomics, its proper application deserves some caution. In this regard, a preliminary assessment of the repeatability of the radiomic features estimation is recommended. Moreover, each step of the radiomic workflow (i.e., image acquisition and reconstruction, image segmentation, image preprocessing, image filtering, and feature extraction) has the potential to influence features estimation. Hence, researchers involved in this field are focusing on the standardization of the radiomic workflow, in order to possibly use radiomic features as imaging biomarkers (IB). Given that CMR radiomics is in its infancy, there is a need to thoroughly evaluate these aspects for each specific application. Accordingly, the purpose of this work was to investigate, for the first time, the repeatability of myocardial radiomic features from quantitative CMR T1 and T2 mapping along with their sensitivity to image preprocessing and image filtering.