Novel spatial interpolation methods for catheter-based mapping of endocardial arrhythmias

Abstract

Due to their minimal invasiveness catheters are highly prefered as cardiac mapping techniques used in the source localization of rhythm disturbances in the heart (arrhythmias). It is known that most of arrhythmias originate from the inner surface of the heart (endocardium). In mapping this surface two catheter-based approaches are used which are called standart and multielectrode basket catheters. In this study in order to reduce the mapping duration it is aimed that acquiring fewer measurements and spatially interpolating the remaining unmeasured sites on the endocardium, and obtaining high-resolution maps using spatial interpolation methods from limited number of measurements acquired by multielectrode basket catheters. For these purposes, three spatial interpolation approaches for different spatial resolution of measurements are investigated. These approaches are spline, Laplacian, and Newton's interpolation methods. In the application of these methods on the abovementioned problems high resolution test maps are created using computer simulation of electrical activity of the heart. The results show that spline and Laplacian methods used in the interpolation of left and right endocardial data performed more successfully than Newton's method. Even when 5% of the points on endocardial surfaces are assumed to be known a mean correlation coefficient of ~0.99 is obtained.