Accessibility is one of key factors in saving life, particularly in case of cardiovascular diseases. However, the implementation of portable ECG devices often leads to misclassifications. This paper aims to perform feature analysis on the data extracted from our portable ECG device. This paper introduces a portable ECG monitoring system with built-in classification capabilities. We also add features on our device to classify normal and abnormal signals using Random Forest with XGBoost. This system has the potential to save time, lives, and costs. Notably, the proposed algorithm achieves an accuracy rate of over 98% in distinguishing between normal and abnormal signals.