Computer-aided detection of lung nodules using outer surface features.

摘要: In this study, a computer-aided detection (CAD) system was developed for the of lung nodules in computed tomography images. The CAD consists four phases, including two-dimensional and three-dimensional preprocessing phases. feature extraction phase, different groups features are extracted from volume interests: morphological features, statistical histogram outer surface, texture surface. support vector machine algorithm is optimized using particle swarm optimization classification. provides 97.37% sensitivity, 86.38% selectivity, 88.97% accuracy 2.7 false positive per scan three classification features. After inclusion surface results reaches 98.03% 87.71% 90.12% 2.45 scan. Experimental demonstrate that nodule candidates useful to increase sensitivity decrease number positives