An early and accurate diagnosis and detection of the coronary heart disease using deep learning and machine learning algorithms
Seda Demir, HARUN SELVİTOPİ, Zülküf Selvitopi
Abstract
Abstract This study provides an extensive analysis of the role of Machine Learning (ML) and Deep Learning (DL) techniques in the early diagnosis of Coronary Heart Disease (CHD), one of the primary causes of cardiovascular morbidity and mortality worldwide. Early diagnosis is crucial to slow disease progression, prevent severe complications such as heart attacks, and enable timely interventions. We examine the impact of dataset variability on model performance by applying various ML and DL algorithms, including Multilayer Perceptron (MLP), Artificial Neural Networks (ANN), Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Support Machine Vector (SVM), Logistic Regression (LR), Decision Tree (DT), k–Nearest Neighbor (kNN), Categorical Naive Bayes (CategoricalNB), and Extreme Gradient Boosting (XGBclassifier) to two distinct datasets: the comprehensive Framingham dataset and the UCI Heart Disease dataset. Before model training, data preprocessing techniques such as Hotdecking, Synthetic Minority Oversampling Technique (SMOTE), and normalization were implemented to enhance data quality. Model performance was evaluated using a range of metrics, including accuracy, precision, recall, F1–score, and area under the curve (AUC). The results reveal that the SVM model achieved the highest accuracy of $$92.42\%$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mn>92.42</mml:mn> <mml:mo>%</mml:mo> </mml:mrow> </mml:math> on the UCI dataset, while XGBclassifier attained the highest accuracy of $$90.97\%$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mn>90.97</mml:mn> <mml:mo>%</mml:mo> </mml:mrow> </mml:math> on the Framingham dataset, surpassing the performance reported in existing literature. These findings emphasize the potential of ML and DL methods for the early diagnosis of CHD and demonstrate the significance of dataset selection on model performance. This study offers valuable insights into the effectiveness of ML and DL approaches, underscoring the importance of data–driven strategies in advancing healthcare for the early detection and management of CHD and similar cardiovascular diseases.