Key result
Random Forest with polygon approximation achieves ~99% accuracy for ECG pattern classification.
Why the study?
A novel approach using polygon approximation techniques for ECG pattern recognition with machine learning classification methods has not been explored.
Combining the Ramer Douglas Peucker polygon approximation technique with a Random Forest classifier achieves 98.5% accuracy in classifying ECG patterns for heart disease.
The identification of patterns in electrocardiograms (ECGs) involves analyzing ECG signals through deep learning (DL) or machine learning (ML) methods to recognize specific patterns associated with various heart conditions. This study focuses on diagnosing heart disease patterns using polygon approximation techniques to extract features from ECG signals, offering a novel approach that hasn’t been explored in the realm of ECG pattern recognition with machine learning classification methods. We evaluate two dominant point techniques. The first technique involves polygon approximation, and the second technique is based on curvature, implemented through the Ramer Douglas Peucker (RDP) algorithm and the Rosenfeld and Johnston (R&J) algorithm, respectively. This evaluation is conducted using two parameters: execution time and classification accuracy. We used several ML algorithms for the classification task, such as XGBoost, Random Forest (RF), Support Vector Machines (SVMs) with different kernels, and K-Nearest Neighbors (KNNs). RF classifier in conjunction with the Ramer Douglas Peucker approximation produced the most significant classification accuracy of 98.5%.
No takes yet. Share an insight, caveat, or question.
Ali et al. (2026) studied Cardiovascular disease (ECG pattern recognition) (n=2,400). Random Forest classifier with Ramer Douglas Peucker (RDP) polygon approximation vs. Rosenfeld and Johnston algorithm and other machine learning classifiers was evaluated on Classification accuracy. The Random Forest classifier in conjunction with the Ramer Douglas Peucker polygon approximation technique achieved the highest ECG pattern classification accuracy of 98.5%.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: