Date of Award
Santa Clara : Santa Clara University, 2019.
Master of Science (MS)
In this study, a novel Atrial Fibrillation (AFib) detection algorithm is presented based on Electrocardiography (ECG) signals. In particular, the spectrogram of ECG signal is used as an input to a Convolutional Neural Network (CNN) to classify normal and AFib ECG signals. This model is shown to perform well with an accuracy of 92.91% and a value of 0.9789 for the area under the ROC curve (AUC). This study demonstrated the potential of using image classification methods and CNN model to detect abnormal biosignals with noise.
Lu, Senbao, "Automated Atrial Fibrillation Detection from Electrocardiogram" (2019). Bioengineering Master's Theses. 5.