Fusing texture, shape and deep model-learned information at decision level for automated classification of lung nodules on chest CT

• Published in: Information fusion Vol. 42; pp. 102 - 110
• Main Authors: Xie, Yutong, Zhang, Jianpeng, Xia, Yong, Fulham, Michael, Zhang, Yanning
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.07.2018
• Subjects: AdaBoost, information fusion; Back propagation neural network (BPNN); Chest CT; Deep convolutional neural network (DCNN); Lung nodule classification; Lung nodule classification; AdaBoost, information fusion; Chest CT; Deep convolutional neural network (DCNN); Back propagation neural network (BPNN)

•Using nodule heterogeneity (texture/shape) features and representation learned by a deep model.•Constructing an ensemble classifier using back propagation neural network and AdaBoost.•Fusing the decisions made by 3 ensemble classifiers, which are trained on 3 features, respectively.•Outperforming t...