|
|
|
AUTHOR(S): Cristinel-Gabriel Rusu, Simona Moldovanu, Luminita Moraru
|
|
TITLE Classification of Cardiovascular Diseases using ECG Signals and a Genetic Algorithm |
 PDF |
|
ABSTRACT To assist physicians in diagnosing cardiovascular diseases (CVD), a variety of meaningful features such as QRS complex, T and JJ waves, QRS and QRST areas of ECG signal were extracted and further, they fed seven machine learning algorithms for CVD classification. The initially selected features are optimized using a Genetic Algorithm (GA) optimization algorithm and the new features are fed the machine learning classifiers. Thus, traditional machine learning Decision Tree (DT), Random Forest (RF), Ada Boost (AB), Quadratic Discriminant Analysis (QDA), Gaussian NB (GNB), K Nearest Neighbors (KNN), and Gradient Boosting (GB) techniques were implemented to the ECG signals. The same classification task is performed without a GA optimization algorithm. The Kaggle database of electrocardiograms (ECG) containing patients with and without cardiovascular diseases was used to classify ECG signals. The best classification performance results were collected when the machine learning algorithms were optimized using GA. |
|
KEYWORDS ECG signal classification, genetic algorithm, machine learning classifiers, optimization |
|
|
|
Cite this paper Cristinel-Gabriel Rusu, Simona Moldovanu, Luminita Moraru. (2024) Classification of Cardiovascular Diseases using ECG Signals and a Genetic Algorithm. International Journal of Biology and Biomedicine, 9, 1-6 |
|
|
