What question did this study set out to answer?

The aim is to develop an automated classifier for diagnosing multiple classes of heart disease from ECG images.

March 30, 2026Open Access

Multi-class Heart Disease Classification Using Multi-lead ECG Features and Ensemble Learning

Key Result

An automated ensemble learning model using multi-lead ECG image features achieved 98.06% accuracy in classifying four cardiac conditions, outperforming existing models.

Key Points

The aim is to develop an automated classifier for diagnosing multiple classes of heart disease from ECG images.
Utilized a preprocessing pipeline including lead segmentation and grayscale conversion.
Extracted contour-based features from ECG images, followed by dimensionality reduction using PCA.
Employed ensemble learning with models such as KNN, logistic regression, and SVM using voting and stacking methods.
Achieved a classification accuracy of 98.06% on the dataset.
Outperformed existing pre-trained models and state-of-the-art approaches.
Successfully categorized ECG images into four classes: normal, abnormal, myocardial infarction, and history of MI.

Structured PICO

Population

Public dataset of ECG images categorized into four classes: normal, abnormal, myocardial infarction (MI), and history of MI

Intervention

Automated multiclass classifier using image processing and ensemble machine learning (voting and stacking classifiers of KNN, logistic regression, and SVM)

Comparator

Existing pre-trained and state-of-the-art models

Outcome

Classification accuracy

An automated ensemble machine learning framework achieved 98.06% accuracy in classifying ECG images into normal, abnormal, MI, and history of MI categories.

Main Result

Absolute Event Rate: 0% vs 0%

Abstract

Cardiovascular diseases (CVDs) are the leading causes of global mortality and require an early and precise diagnosis. This work presents an automated multiclass classifier for diagnosing cardiac disease from electrocardiogram (ECG) images through image processing and machine learning techniques. The proposed framework consists of three steps, including preprocessing, feature extraction, and ensemble learning. Initially, the ECG image undergoes a comprehensive preprocessing pipeline that includes lead segmentation, grayscale conversion, Gaussian filtering, and Otsu thresholding. The contour-based features are extracted and then reduced by PCA to preserve discriminative information. Finally, multiple machine learning models, including K-nearest neighbors (KNN), logistic regression, and support vector machines (SVM), are ensembled using voting and stacking classifiers to improve the performance of the proposed framework. The proposed ensemble model is evaluated on a public dataset that consists of ECG images that are categorized into four classes: normal, abnormal, myocardial infarction (MI), and history of MI. The proposed ensemble model attained the highest classification accuracy of 98.06% and outperforms the existing pre-trained and state-of-the-art models.