Development of alternative diagnostic feature system in the cardiology decision support systems

Abstract

The trend towards an increase in the production of Ukrainian digital electrocardiographic telemetry systems such as transtelephonic digital 12-channel electrocardiograph complex "Telecard" identified the need to create intelligent automated cardiac decision support systems. The basis of these systems is the morphologic analysis of electrocardiograms, which represent biomedical signals with locally concentrated features. The system of alternative diagnostic features based on the method proposed by the authors of the morphological analysis of biomedical signals with locally concentrated features to provide additional graphical information in the diagnosis of one of the most common cardiac arrhythmias - ventricular arrhythmia is developed. Representation of the electrocardiogram in two-dimensional space of alternative features, as well as hodograph is proposed. Differences between the ECG-hodographs for normal ECG and ECG with different arrhythmias of right and left ventricles, as well as multifocal ventricular arrhythmia are analyzed. It was found that a graphical representation of an electrocardiogram in the alternative feature space allows the physician to visually perform the classification of different types of ventricular arrhythmia, which in combination with the classical analysis of ECG on the time axis increases the reliability of diagnostics.

Разработана система альтернативных диагностических признаков на основе метода морфологического анализа биомедицинских сигналов с локально сосредоточенными признаками с целью предоставления дополнительной информации при диагностике желудочковой экстрасистолии. Предложено представление электрокардиограммы в альтернативном пространстве признаков в виде годографа. Проанализированы отличия годографов для нормальной электрокардиограммы и электрокардиограмм с различными видами желудочковой экстрасистолии.