Geometry poincare-characterized cardiac disease identification scheme in physio-grid

Abstract

Recently several Grid systems, especially Physio-Grid, have been studied and proposed for advanced physiological disease identification. The Physio-Grid system has been designed to support advanced physiological disease identification with data integration of distributed medical database in which electrocardiogram (ECG) signals. Advances in patient care and monitoring have allowed physicians to track a patient’s physiological state more closely with high accuracy. Modern handheld healthcare systems can take information from multiple data sources and store them in a healthcare server. The data collected from patient monitors and clinical information systems must be indexed and presented in a user-friendly interface. However, the sophisticated disease identification schemes for accurate detection and diagnosis of cardiac diseases have not reported well in literatures. In order to develop advanced cardiac diseases using ECG analysis, we examined several methods such as mental stress analysis with heart rate, heart rate variability (HRV) analysis, and geometric Poincare plot analysis. In this thesis, we propose a new scheme for cardiac disease identification applicable for Physio-Grid system. We calculated Poincare plot descriptors such as mean square and standard deviation. The Poincare descriptors are evaluated using two kinds of ECG signal data from PhysioBank. We detected that the descriptors calculated from the arrhythmia ECG database are different to the descriptors calculated from the long-term ECG database. From the experimental result, we propose a geometry Poincare characterized cardiac disease identification scheme for being implemented and advanced Physio-Grid.