Anomaly Detection in ECG Signal Using Unsupervised Learning

An essential diagnostic tool for finding irregularities in the rhythm and function of the heart is the electrocardiogram (ECG) signal. After obtaining the ECG signals from the adjacent hospital, we preprocessed and examined the data. We used unsupervised learning methods for anomaly detection in this paper. Important elements of the ECG signal are also covered in detail in our study. With the goal of increasing the precision and efficacy of cardiac abnormality diagnosis, we used this knowledge to conduct research on anomaly detection using unsupervised machine learning algorithms. After comparison of algorithm for anomaly detection, we found that autoencoder performance was best than other algorithms. This study is in line with our goal of advancing innovative technology that improves diagnostic accuracy while also fostering sustainable health solutions, which in turn strengthens the resilience of medical infrastructure and promotes long-term well-being to meet the SDGs.