Detection and Classification of Arrhythmia using Machine Learning

Arrhythmias, which occur when the heart beats too fast, too slowly, or irregularly, continue to pose a serious challenge to global public health. Electrocardiogram (ECG) analysis remains the primary diagnostic tool for identifying these rhythm abnormalities; however, manual interpretation by clinicians is often time- consuming and susceptible to human error. To address this limitation, the present study proposes an intelligent, machine learning–based framework for the automatic detection and classification of arrhythmias using ECG data. The approach incorporates signal preprocessing, feature extraction, and classification through the Weighted K-Nearest Neighbors (WKNN) algorithm to achieve consistent and reliable performance. Experimental evaluation demonstrates that the proposed system delivers high diagnostic accuracy and robust adaptability, making it well-suited for integration into clinical decision-support tools and wearable health-monitoring devices.