FYP2020 | Ventricular Arrhythmia ECG Analysis Using Discriminant Analysis And Support Vector Machine скачать в хорошем качестве

FYP2020 | Ventricular Arrhythmia ECG Analysis Using Discriminant Analysis And Support Vector Machine

This project proposed the comparison between two classification techniques which are Discriminant Analysis (DA) and Support Vector Machine (SVM). Normal (N) and ventricular arrhythmia (V) ECGs will be classified using the features extracted from the signals which are mean, standard deviation and root mean square (RMS) of R-R interval, QRS complex and R peak. 17 Normal ECG signals are taken from MIT-BIH Normal Sinus Rhythm and 35 ventricular arrhythmia ECG signals are taken from CU Ventricular Tachyarrhythmia Database. The performance of both classifiers will be compared in terms of accuracy, sensitivity, specificity and precision where the results of DA are 88.23%, 87.44%, 90.2% and 94.85% respectively while SVM are 85.51%, 85.87%, 84.31% and 92.68% respectively. It is proved that DA has higher performance than SVM in classifying normal rhythm from ventricular arrhythmia rhythm. #DiscriminantAnalysis #SupportVectorMachine Contact Supervisor: Dr. Mohd Afzan Bin Othman Department of / Centre of Faculty of Electrical Engineering Universiti Teknologi Malaysia Johor Bahru, Malaysia. [email protected]