1D Convolutional Neural Network to Detect Ventricular Fibrillation

Ventricular fibrillation (VF) contributes to the majority of arrhythmia mortality and morbidity rate, as studies show the survival rate of patients who have been discharged from the hospital less than forty percent, while the mortality rate of patients who did not have fast access to defibrillator exceeds 90 percent. The research aims to develop another method of ventricular fibrillation detection. The research utilizes a convolutional neural network (CNN) with ten-second ECG data gathered from PhysioNet database CU Ventricular Tachyarrhythmia Database (CUDB) to determine ventricular fibrillation reading from normal reading. An accuracy of 90%, a sensitivity of 96%, and a specificity of 84% of test data were obtained. In conclusion, the CNN can be possibly implemented for the VF early detection system.