As the heart, and particularly the left ventricle (LV), tends to remodel its shape under varying pathological conditions, it is hypothesized that the quantitative assessment of its three-dimensional (3-D) geometry may help in the objective discrimination between different pathological conditions. The 3-D LV geometries of 27 human subjects were studied using ultrafast tomography (Cine-CT) cross-sectional multislice imaging. Ten of the studied hearts were normal (NR); nine had LV aneurysms (AN), five had myocardial infarctions (MI), and three had hypertension/hypertrophic hearts (HT). After tracing the endocardial boundaries, the end systolic (ES) shape of each LV was represented by its corresponding “geometrical cardiogram’’ (GCG), a recently introduced shape descriptor used to characterize the LV’s 3-D instantaneous geometry by an anatomically aligned normalized helical vector. Applying a Fourier-sine series expansion to the GCG yielded a geometrical spectrum, which decomposed the LV shape into a family of sinusoidal geometries. It was found that the different pathological states of the LV were associated with characteristic changes in the geometrical spectrum domain. Representing each heart by a feature vector in the spectral domain and applying unsupervised fuzzy clustering to the obtained 27 feature vectors, an overall success of 85% in classification was obtained. These results indicate that an operator-independent shape-based diagnosis is potentially feasible for the four different pathological categories studied here.
ASJC Scopus subject areas
- Computer Science Applications
- Electrical and Electronic Engineering
- Radiological and Ultrasound Technology
- Computational Theory and Mathematics
- Biomedical Engineering
- Radiology Nuclear Medicine and imaging