Quantitative detection of left ventricular dyssynchrony from cardiac computed tomography angiography

Heart failure represents both an enormous disease and economic burden in the United States, affecting more than 5.7 million people nationally at an annual cost of more than 35 billion dollars. Cardiac computed tomography angiography (CCTA) is a rapidly advancing, non-invasive imaging technique that has the potential to both dramatically reduce the cost and simplify the gathering of diagnostic information needed for the evaluation and treatment of patients with newly diagnosed LV systolic dysfunction. In this paper, we present a parametric model based approach for the detection and classification of dyssynchronous LV using CCTA data. First, the LV endocardial border is traced and then fitted in a prolate spheroidal coordinate system. A new metric, time to minimum λ (t_min λ), is then derived from this parametric model and tested on six LVs (3 synchronous, 3 dyssynchronous). The preliminary results of the classifier using only the means and standard deviations of t_min λ are extremely encouraging. We then further show how t_min λ can be used to quantitatively study the degree and location of dysfunction in a dyssynchronous LV.