Analyzing and Classification of Heart Disease using VGG19

This study presents a novel method for heart disease classification by integrating the VGG 19 architecture with a 2D Convolutional Neural Network (CNN). This approach aims to improve diagnostic accuracy and reduce misdiagnosis risks. It involves preprocessing heart images, selecting relevant features, and applying classification algorithms. The method achieved a $\mathbf{9 0 \%}$ accuracy rate in data validation. It uses CNNs for detailed data analysis and comparison, and introduces a hybrid technique combining VGG 19 with a 2D CNN. Various classification models were tested, with the GLCM combined with GoogleNet emerging as the most effective for feature selection and accurate heart disease prediction.