MRI Post-Processing Methods for Myocardial Infarct Quantification

Myocardial infarct (MI) size has been increasingly used as an endpoint in multiple clinical trials and has thus become an important clinical measure. While late gadolinium enhancement MRI is considered the clinical reference standard to detect, characterize, and quantify MI, there is no established universal quantification algorithm that provides reliable MI assessment in every scenario. Efforts have been made to improve the binary threshold-based methods which dichotomize MRI voxels as either healthy or infarcted. Novel algorithms have also been proposed to quantify the actual infarcted tissue content of each MRI voxel while accounting for partial volume averaging, a common issue in quantitative MRI. Currently, the full-width at half-maximum binary algorithm seems to have the highest accuracy and reproducibility. Non-binary algorithms show comparable results; however, the literature is limited in terms of their clinical feasibility.