Performance Analysis of Coronary Artery Angiography Detection using Hybrid Deep Learning Algorithm

This contribution explores the use of heart angiography images to distinguish between healthy and coronary artery conditions using a newly developed deep learning approach. The method involves three key components: a transformation model, a feature extraction process, and an image classification algorithm. The NSCT algorithm is applied to transform the angiography images into a set of non-linear coefficients, which serve as the foundation for feature extraction. Unique features are then derived from both healthy and coronary artery images. A hybrid Convolution Neural Network (CNN) is introduced, which integrates both machine learning and deep learning techniques to classify the angiography images with high accuracy. This approach has been thoroughly tested on two different heart angiography datasets and has shown promising results in correctly identifying coronary artery conditions. The performance of the proposed method is also compared with other recent techniques in the field to highlight its effectiveness.