Early detection of heart rhythm disorders is important for preventing serious cardiac problems. Arrhythmia, an irregular heartbeat caused by abnormal electrical activity, often goes unnoticed without proper monitoring. This project develops an arrhythmia detection system using machine learning applied to electrocardiogram (ECG) data. The system analyses ECG waveform data where each heartbeat is represented by sampled signal values that capture the heart’s electrical activity. By studying these patterns, the model can distinguish between normal and abnormal heartbeats. A machine learning model is trained on labelled ECG data to predict the probability of arrhythmia for each heartbeat. The trained model is integrated into a web-based application that allows users to input ECG data and receive real-time arrhythmia risk predictions. This project demonstrates how machine learning and ECG analysis can assist healthcare professionals in early diagnosis and continuous heart monitoring.