ECG Feature Elements Identification For Cardiologist Expert Diagnosis

This paper proposes a reliable method for cardiologist expert diagnosis based on ECG elements identification. This method analyzes ECG key features (P wave, QRS complex, T wave). It includes noise purification, sample design for digital ECG, understanding of the HP ECG criteria program, and the extended measurements report. This project report synthesizes the advantages of math, multiple function analysis, database and knowledge base, and expert system to explore the mechanism of "ECG feature elements identification for cardiologist expert diagnosis". This report proposes an integral method of ECG information flow for its area computing and a differential method of ECG information flow for its slope computing and a convolution method for true ECG wave form element extraction even they confuse with each other or incomplete. This method can implement ECG report in real time and provide exact explanations for the diagnostic decision obtained. This method can offer mean (standard) values estimation for parameters and confident interval computing for predictive accuracy (above 85%). This method solved the following problems: noise purification, incomplete and confusing ECG element\´s key features identification, decision-rule base, and expert diagnosis model - for "cardiologist expert diagnosis" research topic