Bölüm Künyesi
Bölüm Özeti
<br>Cardiovascular disease (CVD) continues to be one of the leading causes of death worldwide. Early diagnosis plays a critical role in preventing fatal outcomes related to CVD. To achieve this, data analysis is essential in identifying potential risks and intervening in a timely manner. In this study, we explore the application of machine learning techniques to aid in the early detection and diagnosis of cardiovascular diseases. A hybrid dataset named Sathvi, which integrates the Hungarian, Switzerland, Cleveland, and Long Beach datasets, has been analyzed using deep learning techniques, convolutional neural network (CNN). The "Hybrid" and "Sathvi" datasets, with 920 and 531 instances respectively. The "Hybrid" dataset includes 920 instances, but over 50% of the values for the 'ca' and 'thal' features are missing. To address this, these two columns were removed, resulting in the creation of a new dataset called "Sathvi." Furthermore, any instances with missing values in the "Hybrid" dataset were eliminated. As a result, the final "Sathvi" dataset contains 531 instances and 12 features, all without missing data. The performance of the models was assessed using metrics including Accuracy, Recall, F1 Score, and AUC (Area Under the Curve). Obtained performance metrics compared with Logistic Regression (LR), k-nearest neighbour (k-NN) and Random Forest Classifier(RFC) machine learning techniques.