선형 예측 계수와 Fuzzy ARTMAP을 이용한 심전도 패턴 인식 = A Neural Network(Fuzzy ARTMAP) Approach for ECG Pattern Recognition Using LPC Coefficients

다국어 초록 (Multilingual Abstract)

I have investigated the QRs complex, extracted from ECG electrocardiogram data, using fuzzy ARTMAP(adaptive resonance theory mapping) to classify cardiacqr arrhythmias. The ECG signals used for the purpose of evaluation in this study were taken from the MIT/BIH arrhythmia database. Two different conditions have been analyzed, normal and the abnormal PVC(prematrure ventricular con traction). Based on the MIT/BIH database annotations, cardiac beats for normal and abnormal QRS complexes were extracted, scaled, and Hamming windowed, after bandpass filtering, to yield a sequence of 100 samples for each QRS segment. From this sequence, two LPG(linear predictive coding) coefficients were generated using the Burg's maximum entropy method. The two LPG coefficients, along with the mean-square value of the QRS complex segment, were utilized as fratures of each condition. Using these three time series parameters, the fuzzy ARTMAP neural network was trained and tested for the classification of normal and abnormal PVC conditions. The test results show that the fuzzy ARTMAP neural network can classify cardiac arrhythmias with greater than 99% specificity, and 97% sensitivity.

목차 (Table of Contents)

• Ⅰ. INTRODUCTION
• Ⅱ. METHOD OVERVIEW
• Ⅲ. ECG PREPROCESSING
• Ⅳ. FUZZY ARTMAP ARCHITECTURE
• Ⅴ. EXPERIMENTAL RESULTS AND DISCUSSIONS
• Ⅵ. CONCLUSIONS
• REFERENCES