Cost sensitive adaptive random subspace ensemble for computer-aided nodule detection

Many lung nodule computer-aided detection methods have been proposed to help radiologists in their decision making. Because high sensitivity is essential in the candidate identification stage, there are countless false positives produced by the initial suspect nodule generation process, giving more work to radiologists. The difficulty of false positive reduction lies in the variation of the appearances of the potential nodules, and the imbalance distribution between the amount of nodule and non-nodule candidates in the dataset. To solve these challenges, we extend the random subspace method to a novel Cost Sensitive Adaptive Random Subspace ensemble (CSARS), so as to increase the diversity among the components and overcome imbalanced data classification. Experimental results show the effectiveness of the proposed method in terms of G-mean and AUC in comparison with commonly used methods.

[1]  Qiang Li,et al.  Recent progress in computer-aided diagnosis of lung nodules on thin-section CT , 2007, Comput. Medical Imaging Graph..

[2]  Hiroyuki Yoshida,et al.  Three-dimensional computer-aided diagnosis scheme for detection of colonic polyps , 2001, IEEE Transactions on Medical Imaging.

[3]  Giorgio Valentini,et al.  Support vector machines for candidate nodules classification , 2005, Neurocomputing.

[4]  Nitesh V. Chawla,et al.  Editorial: special issue on learning from imbalanced data sets , 2004, SKDD.

[5]  Qiang Li,et al.  Selective enhancement filters for nodules, vessels, and airway walls in two- and three-dimensional CT scans. , 2003, Medical physics.

[6]  Michael R Hamblin,et al.  CA : A Cancer Journal for Clinicians , 2011 .

[7]  Jacek M. Zurada,et al.  Training neural network classifiers for medical decision making: The effects of imbalanced datasets on classification performance , 2008, Neural Networks.

[8]  Victor S. Sheng,et al.  Thresholding for Making Classifiers Cost-sensitive , 2006, AAAI.

[9]  I. Tomek,et al.  Two Modifications of CNN , 1976 .

[10]  Anil K. Jain,et al.  Learning-based pulmonary nodule detection from multislice CT data , 2004, CARS.

[11]  Salvatore J. Stolfo,et al.  AdaCost: Misclassification Cost-Sensitive Boosting , 1999, ICML.

[12]  Tin Kam Ho,et al.  The Random Subspace Method for Constructing Decision Forests , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[13]  S. Armato,et al.  Massive training artificial neural network (MTANN) for reduction of false positives in computerized detection of lung nodules in low-dose computed tomography. , 2003, Medical physics.

[14]  K. Doi,et al.  Computerized detection of lung nodules in thin-section CT images by use of selective enhancement filters and an automated rule-based classifier. , 2008, Academic Radiology.

[15]  Haibo He,et al.  Learning from Imbalanced Data , 2009, IEEE Transactions on Knowledge and Data Engineering.

[16]  N. Dubrawsky.  Cancer statistics , 1989, CA: a cancer journal for clinicians.

[17]  Nitesh V. Chawla,et al.  SMOTEBoost: Improving Prediction of the Minority Class in Boosting , 2003, PKDD.

[18]  Nitesh V. Chawla,et al.  SMOTE: Synthetic Minority Over-sampling Technique , 2002, J. Artif. Intell. Res..

[19]  Lilla Böröczky,et al.  Feature Subset Selection for Improving the Performance of False Positive Reduction in Lung Nodule CAD , 2005, IEEE Transactions on Information Technology in Biomedicine.

[20]  Taylor Murray,et al.  Cancer statistics, 2000 , 2000, CA: a cancer journal for clinicians.

[21]  Masayuki Murakami,et al.  Adaptive filter to detect rounded convex regions: iris filter , 1996, Proceedings of 13th International Conference on Pattern Recognition.