Exploratory treatment of inverse problems

In this overview the authors have consider variance-based models of epicardial and body surface distributions, but the relationships between mean vectors, as well as higher order moments of each distribution, can be modeled as well. For example, the fourth moment has been shown to be important and can be considered in Karhunen-Loeve transformations. Also, consideration must be given to the minimum sample size needed to obtain a good estimate of the population. The authors emphasize that the approaches outlined here are an exploratory treatment of the inverse problem, they are not necessarily a means to solve it. Additionally, one cannot, for example, infer the presence of individual physiologic generators from the principal components. Rather, the motivation of this work is to develop a better understanding of spatiotemporal relationships pertaining to the observation and source spaces, through exploratory analyses