Exploring Diverse Machine Learning Approaches for Coronary CT Images Analysis: A Comparative Study

This study evaluates the performance of various machine learning algorithms on coronary computed tomography angiography (CCTA) images for detecting coronary atherosclerosis. The algorithms include Logistic Regression (LR), Naïve Bayes (NB), K-Nearest Neighbors (KNN), Decision Trees (DT), Random Forest (RF), Support Vector Machines (SVM), Multi-Layer Perceptron (MLP), Recurrent Neural Network (RNN), Convolutional Neural Network (CNN), and Convolutional Neural Network with Transfer Learning (CNN w/ TL). Performance metrics such as Accuracy, Recall, F1-Score, and Area Under the Curve (AUC) are used to compare their effectiveness on separate validation and test datasets. The results highlight $\mathrm{CNN} \mathbf{w / T L}$ as the most effective algorithm, achieving an AUC of $\mathbf{0. 9 1}$ on the validation dataset. The CNN models also perform well, with all metrics over 0.80. Random Forest (RF) demonstrates robust performance with metrics at 0.75 (Validation) and 0.70 (Test).