Cardiovascular disorder severity analysis in magnetic resonance images using texture features and metaheuristic classification algorithms

Abstract

Heart Failure (HF) in human is one of the complex clinical cardiac syndromes that inhibit the heart from satisfying the circulatory demands of the whole body. It may occur due to spectrum of cardiovascular disorders such as myocardial ischemia, myocardial infarction, cardiomyopathy and valvular diseases. Systolic Heart Failure (SHF) and Diastolic Heart Failure (DHF) are the two major subtypes of HF that predominantly exhibit reduced and hyperdynamic Ejection Fraction (EF). It is characterized by external symptoms such as shortness of breath, fatigue and dyspnea. Currently, the HF stages such as mild HF, moderate HF, severe HF and hyperdynamic HF are clinically diagnosed using EF or heart rate variability. However, assessment of changes in cardiac tissue structure and functions provide complementary and useful information about the progression of HF. Various researches suggest that HF subtypes share a number of equally severe cardiac dysfunctions, but the pathophysiology has not yet been addressed in a comprehensive way. newlineThe study of Left Ventricle (LV) chamber of heart plays an important role in quantification of cardiac function that is essential to manage cardiac pathologies. Cine Cardiovascular Magnetic Resonance (CMR) sequences are non-invasive, precise and widely considered gold standard for functional and morphological assessment of LV. The LV blood pool area undergoes significant transformation at End Systole (ES) and End Diastole (ED) phase for systolic and diastolic dysfunction respectively. However, this transformation induces geometrical and tissue pattern variations in LV myocardium at ES and ED frames. Hence, the analysis of LV myocardium is the primary prerequisite in HF risk stratification. newline