Radiomics-Based Classification in Cardiovascular Magnetic Resonance
Generally morphological and functional traits increase the diagnosis complexity. Additional clinical information, besides imaging data such as cardiovascular magnetic resonance, is also usually required to reach a definitive diagnosis, including electrocardiography, family history, and genetics. In this study, a total of 118 subjects, including 35 patients with Left Ventricular Non-compaction (LVNC), 25 with Hypertrophic Cardiomyopathy (HCM), 37 with Dilated Cardiomyopathy (DCM), as well as 21 healthy volunteers (NOR), underwent cardiovascular magnetic resonance imaging to consider a radiomics approach. Such approach was automatically encoding differences in the underlying shape, gray-scale and textural information in the myocardium and its trabeculae that could enhance the capacity to differentiate between these overlapping conditions. The use of radiomics models for the automated diagnosis of LVNC, HCM, and DCM resulted in excellent one-vs.-rest receiver operating characteristic (ROC) area under the curve (AUC) values of 0.95 while generating these results without the need for the delineation of the trabeculae. First-order and texture features resulted to be among the most discriminative features in the obtained radiomics signatures, indicating their added value for quantifying relevant tissue patterns in cardiomyopathy differential diagnosis.