Automated Endocardial Border Detection from Cardiac Magnetic Resonance Images for Quantitative Assesment of Left Ventricular Function

Cardiac magnetic resonance (CMR) is a noninvasive cardiac imaging modality with excellent spatial and contrast resolution that has become the standard reference for left ventricular (LV) size and function, against which other techniques are frequently validated. CMR quantification LV size and function requires detection of endocardial boundaries. Most current techniques use image intensity gradients that require manual corrections and are thus subjective and time-consuming. As a result, clinically, CMR has been mostly used to measure only end-systolic and end-diastolic volumes (ESV, EDV) and calculate ejection fraction (EF). Availability of a reliable technique for fully-automated detection of the endocardial boundaries throughout the cardiac cycle would overcome this limitatio. In addition, LV volume over time curves may provide clinically important information on LV dynamics, and provide direct insight into LV contraction and relaxation properties closely related to pathophysiology of various disease states. Our specific aims were: (1) to determine to what extent the lack of dynamic information affects LV volume and EF measurements; (2) to validate the automatically detected endocardial boundaries as well as calculated cross-sectional LV areas and volumes against conventional methodology based on manual tracing in patients with a wide range of ejection fractions; and (3) to test clinical feasibility of the former approach by applying it to images obtained in patients with abnormal LV systolic and diastolic function.