Automatic detection and diagnosis of lung nodules on CT images based on LDA and SVM / 南方医科大学学报

ABSTRACT

Based on suspected pulmonary nodule segmentation images obtained previously and with a large-sample training, automatic detection and diagnosis of the pulmonary nodules on CT images was realized by extracting the multi-dimensional features of the pulmonary nodule images and the application of LDA and SVM statistical classifiers. Experimental results showed that this detection and diagnosis method produced better classification results, and is practical for application in CAD systems.