Abstract: Automated Electrocardiogram (ECG) classification and arrhythmia detection represent a cutting-edge frontier in cardiac diagnostics, poised to revolutionize the identification and management of cardiovascular disorders. Leveraging advanced machine learning methodologies, this report delves into the development and implementation of an automated ECG analysis system. Specifically, the integration of the state-of-the-art EfficientNetB7 model and the versatile Random Forest classifier is investigated for its efficacy in discerning complex patterns within ECG signals and categorizing them into distinct arrhythmia classes.