Symptomatic and Asymptomatic Carotid Plaques Classification using CT Images and Hybrid Deep Transfer Learning

This paper proposes a data-driven model for classifying symptomatic and asymptomatic carotid plaques using Computed Tomography (CT) images. We present a hybrid deep transfer learning framework that combines CNN architectures for feature extraction. The extracted features effectively capture complex local and global textures, as well as morphological characteristics, of the carotid artery. These features were subsequently used as inputs for various machine learning models, whose performance was rigorously evaluated on real-world data. The results demonstrate the feasibility and effectiveness of the proposed method in classifying carotid plaques and highlight its potential to improve clinical diagnosis and risk assessment.