Temporal and spatial regularization for imaging cardiac electrical function

The noninvasive imaging of cardiac electrical function attains more interest during the last years because of the availability of multichannel-SQUID (Superconducting QUantum Interference Devices) magnetometer and body surface potential mapping systems. The determination of the electrical sources in the human heart from simulated and measured body surface potential (BSP) and magnetocardiogram (MCG) data is object of the project applied for. Currently most of the literature deals with the localization of focal events at one time instant neglecting information from temporal continuity. These algorithms can be used for the determination of focal sources. When the depolarization (or repolarization) wave is distributed over a widespread area than the traditional inverse approachs (e.g. the minimum-norm algorithm) are not able to describe properly the distributed source pattern. Therefore more powerful inverse algorithms including temporal and spatial constraints have to be considered. The objective of this project is the development of an inverse algorithm for the estimation of the activation time (AT), of the action potential duration (APD) and of the action potential amplitude (APA) map (based on a simplified analytical parameterized model of the AP) including an inhomogeneous boundary element (BE) model of the human torso. The BE thorax model was developed by our group during the last years. Electrical anisotropy will be neglected in the inverse problem for the estimation of the AT, the APD and APA map. In contrast to this the heart`s movement will be considered. The estimation of the AT map (phase 0), of the APD map (phase 3) and of the APA map should enable to image the spatial and temporal pattern of the surface transmembrane potential on the endocardium and epicardium. The applied analytical description for phase O is based on an analytical function and includes the following parameters: time delay. rise time, APA and resting membrane potential. Each of this parameters exepted time delay can be assumed to be a constant or can be estimated at the nodes of the heart`s surface BE structure using nonlinear optimization (modified Gauß-Newton or Levenberg-Marquardt algorithm). A similar description (with a modified rise time) is used for the process of repolarization (phase 3). The surface AT map is of specific interest for the determination of normal or abnormal depolarization pattern in the myocardium. The surface APD map is of interest for the investigation of abnormal repolarization phenomena. This APD map would allow to describe the degree of heterogeneity of the process of repolarization which is also an indicator for the inition of abnormal cardiac propagation. This new type of parameterized inverse algorithm combined with imaging facilities like MRI or CT - for a combination of morphological information and the estimated inverse solution - allows to create a new functional diagnostic tool for the localization of normal and pathological phenomena. Thereby this functional imaging will permit to estimate fully noninvasive the position of arrhythmogenic substrates like ischemic or infarcted areas. accessory pathways. sites of pre-excitation, a quantification of the process of depolarization (AT) and a determination of the pattern of the APD on the heart`s surface.