Bagging Support Vector Machine Approaches for Pulmonary Nodule Detection

Tartar A. , Kilic N., Akan A.

International Conference on Control, Decision and Information Technologies (CoDIT), Hammamet, Tunisia, 6 - 08 May 2013, pp.47-50 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume:
  • Doi Number: 10.1109/codit.2013.6689518
  • City: Hammamet
  • Country: Tunisia
  • Page Numbers: pp.47-50


In this paper, pulmonary nodules extracted from computed tomography (CT) images are classified by the single and bagging support vector machine (SVM) classifiers. To determine features, two dimensional principal component analysis is performed. In order to select the best features, three different models are proposed. These models are tested with classifiers of both single SVM and bagging SVM. As a result of tests, bagging SVM is shown to be superior to single SVM.