Cycles of ECG parameter evolution during ischemic episodes

Presents a comparison between two different methods for analyzing ischemic episodes. Both methods are based on a beat by beat analysis of a large number of parameters. The authors process these data in order to extract information regarding the detection of meaningful cycles (presumably related to ischemic episodes) on ECG parameter evolution. In order to facilitate visual inspection of parameter evolution and thus facilitate the definition of criteria to detect such meaningful cycles, the authors apply a process for reducing the dimensionality of the parameter vector. To accomplish this task they employ both Principal Component Analysis (PCA) and Self-Organizing Maps (SOM). Considering a two dimensional output of these methods, the authors have observed that coinciding with ischemic episodes, a cycle starts moving away from a normality zone and moving back to that zone when the episode ends. As a preliminary result, the authors present graphic examples of these cycles over some of the records of the European ST-T Database.