Three-dimensional left ventricular segmentation from magnetic resonance imaging for patient-specific modelling purposes

Left Ventricular Noncompaction Detected by Cardiac Magnetic Resonance Screening: A Reexamination of Diagnostic Criteria

In a previous cross-sectional screening study of 5,169 middle and high school students (mean age, 13.1 ± 1.78 yr) in which we estimated the prevalence of high-risk cardiovascular conditions associated with sudden cardiac death, we incidentally detected by cardiac magnetic resonance (CMR) 959 cases (18.6%) of left ventricular noncompaction (LVNC) that met the Petersen diagnostic criterion (noncompaction:compaction ratio >2.3). Short-axis CMR images were available for 511 of these cases (the Short-Axis Study Set). To determine how many of those cases were truly abnormal, we analyzed the short-axis images in terms of LV structural and functional variables and applied 3 published diagnostic criteria besides the Petersen criterion to our findings. The estimated prevalences were 17.5% based on trabeculated LV mass (Jacquier criterion), 7.4% based on trabeculated LV volume (Choi criterion), and 1.3% based on trabeculated LV mass and distribution (Grothoff criterion). Absent longitudinal clinical outcomes data or accepted diagnostic standards, our analysis of the screening data from the Short-Axis Study Set did not definitively differentiate normal from pathologic cases. However, it does suggest that many of the cases might be normal anatomic variants. It also suggests that cases marked by pathologically excessive LV trabeculation, even if asymptomatic, might involve unsustainable physiologic disadvantages that increase the risk of LV dysfunction, pathologic remodeling, arrhythmias, or mural thrombi. These disadvantages may escape detection, particularly in children developing from prepubescence through adolescence. Longitudinal follow-up of suspected LVNC cases to ascertain their natural history and clinical outcome is warranted.

Segmentation of the left ventricular endocardium from magnetic resonance images by using different statistical shape models

We evaluate in this paper different strategies for the construction of a statistical shape model (SSM) of the left ventricle (LV) to be used for segmentation in cardiac magnetic resonance (CMR) images. From a large database of LV surfaces obtained throughout the cardiac cycle from 3D echocardiographic (3DE) LV images, different LV shape models were built by varying the considered phase in the cardiac cycle and the registration procedure employed for surface alignment. Principal component analysis was computed to describe the statistical variability of the SSMs, which were then deformed by applying an active shape model (ASM) approach to segment the LV endocardium in CMR images of 45 patients. Segmentation performance was evaluated by comparing LV volumes derived by ASM segmentation with different SSMs and those obtained by manual tracing, considered as a reference. A high correlation (r(2)>0.92) was found in all cases, with better results when using the SSM models comprising more than one frame of the cardiac cycle.