Classification of Arrhythmia Diseases by the Convolutional Neural Network Method Based on Ecg Images

Abstract

Arrhythmia is a heart disorder that refers to an abnormal heartbeat rhythm. Arrhythmia detection uses an electrocardiogram (ECG) to describe the heart's electrical activity. This research aimed to know the performance of the Convolutional Neural Network method in classifying arrhythmia diseases based on ECG signal images. Several stages were used to classify arrhythmias: the pre-processing data stage, CNN model formation stage, model compiling, training, model testing, and evaluation. The CNN model architecture that is formed involves 7 Convolution Layers, 7 Pooling Layers, 2 Dropout Layers, 2 Dense Layers, and 1 Flatten Layer, as well as ReLu and Softmax activation functions. The input variable in the classification process with CNN is an ECG image. The output variable is the classification of ECG signals into 17 classes, including normal sinus and pacemaker rhythms. The processed data are 1000 images; the division scenario is 750 training data and 250 testing data. The result of arrhythmia's classification based on ECG image testing data using the CNN model shows the levels of Accuracy, Precision, Recall, and F1-score levels are 81%, 80%, 71%, and 73%, respectively, respectively. With the F1-score value as a measurement reference, the CNN model performs well in classifying ECG images