CT METHOD AND APPARATUS FOR CLASSIFICATION OF GROUND-GLASS OPACITY NODULES WITH SMALL SOLID COMPONENTS USING MULTIVIEW IMAGES AND TEXTURE ANALYSIS IN CHEST CT IMAGES

摘要

According to an embodiment of the present invention, a method for classifying a ground-glass opacity nodule (GGN) with a small solid component through a multi-view image and a texture analysis in a chest CT image comprises the steps of: (a) generating a multi-view image from a plurality of training images and target images including divided GGN regions; (b) extracting feature vectors of the GGN region from the generated multi-view image and selecting feature vectors to classify a GGN with a small solid component among the extracted feature vectors; and (c) classifying the GGN with a small solid component based on a machine learning-based classifier using the selected feature vectors.