Artificial Intelligence in Medicine
Volume 50, Issue 1 , Pages 43-53 , September 2010

Computer-aided diagnosis of pulmonary nodules using a two-step approach for feature selection and classifier ensemble construction

  • Michael C. Lee

      Affiliations

    • Philips Research North America, 345 Scarborough Road, Briarcliff Manor, NY 10510-2099, USA
    • Corresponding Author InformationCorresponding author. Tel.: +1 914 945 6047; fax: +1 914 945 6580.
    • ,
  • Lilla Boroczky

      Affiliations

    • Philips Research North America, 345 Scarborough Road, Briarcliff Manor, NY 10510-2099, USA
    • ,
  • Kivilcim Sungur-Stasik

      Affiliations

    • College of Physicians and Surgeons, Columbia University, 630 West 168th Street, P&S 8, Room 503, New York, NY 10032, USA
    • ,
  • Aaron D. Cann

      Affiliations

    • College of Physicians and Surgeons, Columbia University, 630 West 168th Street, P&S 8, Room 503, New York, NY 10032, USA
    • ,
  • Alain C. Borczuk

      Affiliations

    • College of Physicians and Surgeons, Columbia University, 630 West 168th Street, P&S 8, Room 503, New York, NY 10032, USA
    • ,
  • Steven M. Kawut

      Affiliations

    • College of Physicians and Surgeons, Columbia University, 630 West 168th Street, P&S 8, Room 503, New York, NY 10032, USA
    • ,
  • Charles A. Powell

      Affiliations

    • College of Physicians and Surgeons, Columbia University, 630 West 168th Street, P&S 8, Room 503, New York, NY 10032, USA

Received 19 September 2008 ,Revised 4 April 2010 ,Accepted 4 April 2010.

References 

  1. Jemal A, Siegel R, Ward E, Hao YP, Xu JQ, Murray T, et al. Cancer statistics. Ca-a Cancer Journal for Clinicians. 2008;58(March–April):71–96
  2. Gurney JW. Determining the likelihood of malignancy in solitary pulmonary nodules with Bayesian analysis. Part I. Theory. Radiology. 1993;186(February):405–413
  3. Nakamura K, Yoshida H, Engelmann R, MacMahon H, Katsuragawa S, Ishida T, et al. Computerized analysis of the likelihood of malignancy in solitary pulmonary nodules with use of artificial neural networks. Radiology. 2000;214(March):823–830
  4. Way TW, Hadjiiski LM, Sahiner B, Chan HP, Cascade PN, Kazerooni EA, et al. Computer-aided diagnosis of pulmonary nodules on CT scans: Segmentation and classification using 3D active contours. Medical Physics. 2006;33(July):2323–2337
  5. Suzuki K, Li F, Sone S, Doi K. Computer-aided diagnostic scheme for distinction between benign and malignant nodules in thoracic low-dose CT by use of massive training artificial neural network. IEEE Transactions on Medical Imaging. 2005;24(September):1138–1150
  6. Siedlecki W, Sklansky J. A note on genetic algorithms for large-scale feature-selection. Pattern Recognition Letters. 1989;10(November):335–347
  7. Kuncheva LI. Combining pattern classifiers: methods and algorithms. Hoboken, NJ: John Wiley & Sons, Inc.; 2004;
  8. Kittler J, Hatef M, Duin RPW, Matas J. On combining classifiers. IEEE Transactions on Pattern Analysis and Machine Intelligence. 1998;20(March):226–239
  9. Kuncheva LI, Bezdek JC, Duin RPW. Decision templates for multiple classifier fusion: an experimental comparison. Pattern Recognition. 2001;34(February):299–314
  10. Boroczky L, Lee MC, Zhao L, Agnihotri L, Powell CA, Borczuk AC, et al. Computer-aided diagnosis for lung cancer using a classifier ensemble. International Journal of Computer Assisted Radiology and Surgery. 2007;2(June):S362–S364
  11. Cunningham P, Carney J. Diversity versus quality in classification ensembles based on feature selection. In: Lopez de Mantaras R, Plaza E, editors. Lecture Notes in Artificial Intelligence 1810: Machine Learning: ECML 2000 11th European Conference on Machine Learning (May 31–June 2, 2000, Barcelona, Spain). Heidelberg, Germany: Springer; 2000. p. 109–16.
  12. Ho TK. The random subspace method for constructing decision forests. IEEE Transactions on Pattern Analysis and Machine Intelligence. 1998;20(August):832–844
  13. Skurichina M, Duin RPW. Bagging, boosting and the random subspace method for linear classifiers. Pattern Analysis and Applications. 2002;5(June):121–135
  14. Guerra-Salcedo C, Whitley D. Feature selection mechanisms for ensemble creation: a genetic search perspective. In:  Freitas AA editors. AAAI-99 and GECCO-99 Workshop on Data Mining with Evolutionary Algorithms: Research Directions. Menlo Park, CA: AAAI; 1999;p. 13–17
  15. Tsymbal A, Pechenizkiy M, Cunningham P. Diversity in search strategies for ensemble feature selection. Information Fusion. 2005;6:83–98
  16. Tsymbal A, Pechenizkiy M, Cunningham P. Sequential genetic search for ensemble feature selection. In:  Kaelbling LP,  Saffiotti A editor. Proceedings of the Nineteenth International Joint Conference on Artificial Intelligence (July 30–August 5, 2005, Edinburgh, Scotland). Denver, CO: Professional Book Center; 2005;p. 877–882
  17. Tsymbal A, Cunningham P, Pechenizkiy M, Puuronen S. Search strategies for ensemble feature selection in medical diagnostics. In:  Krol M,  Mitra S,  Lee DJ editor. Proceedings of the 16th IEEE Symposium on Computer-Based Medical Systems (June 26–27, 2003, New York, NY). Menlo Park, CA: IEEE; 2003;p. 124–129
  18. Opitz D. Feature selection for ensembles. In:  Hendler J,  Subramanian D editor. Proceedings of the 16th National Conference on Artificial Intelligence (July 18–22, 1999, Orlando, Florida). Menlo Park, CA: AAAI Press; 1999;p. 379–384
  19. Dunne K, Cunningham P, Azuaje F. Solutions to instability problems with sequential wrapper-based approaches to feature selection (technical note). Department of Computer Science, Trinity College, University of Dublin; 2002. Jan. Report No.: TCD-CS-2002-28.
  20. Pranckeviciene E, Baumgartner R, Somorjai R. Using domain knowledge in the random subspace method: application to the classification of biomedical spectra. Multiple Classifier Systems. 2005;3541:336–345
  21. Pranckeviciene E, Somorjai R, Baumgartner R, Jeon MG. Identification of signatures in biomedical spectra using domain knowledge. Artificial Intelligence in Medicine. Nov 2005;35:215–226
  22. Bertoni A, Folgieri R, Valentini G. Feature selection combined with random subspace ensemble for gene expression based diagnosis of malignancy. In:  Apolloni B,  Marinaro M,  Tagliaferri R editor. Biological and Artificial Intelligence Environments: Proceedings of the 15th Italian Workshop on Neural Nets (September 14–17, 2004, Perugia, Italy). Dordrecht, Netherlands: Springer; 2004;p. 29–35
  23. Erasmus JJ, Connolly JE, McAdams HP, Roggli VL. Solitary pulmonary nodules. Part I. Morphologic evaluation for differentiation of benign and malignant lesions. Radiographics. 2000;20(January–February):43–58
  24. Li F, Sone S, Abe H, Macmahon H, Doi K. Malignant versus benign nodules at CT screening for lung cancer: comparison of thin-section CT findings. Radiology. 2004;233(December):793–798
  25. Xu Y, Lee MC, Boroczky L, Cann A, Borczuk AC, Kawut S, et al. Comparison of image features calculated in different dimensions for computer-aided diagnosis of lung nodules. In:  Karssemeijer N,  Giger ML editor. SPIE Medical Imaging 2009: Computer-Aided Diagnosis (February 8–12, 2009, Lake Buena Vista, Florida). Bellingham, WA: SPIE; 2009;
  26. Wiemker R, Rogalla P, Blaffert T, Sifri D, Hay O, Shah E, et al. Aspects of computer-aided detection (CAD) and volumetry of pulmonary nodules using multislice CT. British Journal of Radiology. 2005;78:S46–S56
  27. Wiemker R, Rogalla P, Hein P, Blaffert T, Rosch P. Computer-aided segmentation of pulmonary nodules: automated vasculature cutoff in thin- and thick-slice CT. In: Lemke HU, Vannier MW, Inamura K, Doi K, Reiber JHC, editors. International Congress Series. CARS 2003. Computer Assisted Radiology and Surgery. Proceedings of the 17th International Congress and Exhibition (June 25–28, 2003, London UK), 1256; 2003. p. 965–70. doi:10.1016/S0531-5131(03)00283-8.
  28. Hu M. Visual pattern recognition by moment invariants. IEEE Transactions on Information Theory. 1962;8:179–187
  29. Galvez JM, Canton M. Normalization and shape-recognition of 3-dimensional objects by 3D moments. Pattern Recognition. 1993;26(May):667–681
  30. Granlund GH. Fourier preprocessing for hand print character recognition. IEEE Transactions on Computers. 1972;C 21:195–201
  31. Kido S, Kuriyama K, Higashiyama M, Kasugai T, Kuroda C. Fractal analysis of small peripheral pulmonary nodules in thin-section CT—evaluation of the lung-nodule interfaces. Journal of Computer Assisted Tomography. 2002;26(July–August):573–578
  32. Sahiner B, Chan HP, Petrick N, Helvie MA, Goodsitt MM. Computerized characterization of masses on mammograms: the rubber band straightening transform and texture analysis. Medical Physics. 1998;25(April):516–526
  33. Haralick RM, Shanmuga K, Dinstein I. Textural features for image classification. IEEE Transactions on Systems Man and Cybernetics. 1973;SMC-3:610–621
  34. Amadasun M, King R. Textural features corresponding to textural properties. IEEE Transactions on Systems Man and Cybernetics. 1989;19(September–October):1264–1274
  35. Chen CC, Daponte JS, Fox MD. Fractal feature analysis and classification in medical imaging. IEEE Transactions on Medical Imaging. 1989;8(June):133–142
  36. Bishop CM. Neural networks for pattern recognition. New York: Oxford University Press; 1995;
  37. Eshelman L. The CHC adaptive search algorithm: how to have a safe search when engaging in nontraditional genetic recombination. In: Spitz, Bruce M, editor. Foundations of Genetic Algorithms (July 15–18, 1990, Indiana University, Bloomington, Indiana). San Mateo, CA: Morgan Kaufmann.
  38. Guerra-Salcedo C, Whitley DL. Genetic search for feature subset selection: a comparison between CHC and GENESIS. In: Koza, John R, Banzhaf, Wolfgang, Chellapilla, Kumar, Deb, Kalyanmoym Dorigo, Marco, Fogel, David B, Garzon, Max H, Goldberg, David E, Iba, Hitoshi, Riolo, Rick L, editors. Genetic Programming 1998: Proceedings of the Third Annual Conference (July 22–25, 1998, University of Wisconsin, Madison, Wisconsin). San Francisco, CA: Morgan Kaufmann.
  39. Lee MC, Nelson SJ. Supervised pattern recognition for the prediction of contrast-enhancement appearance in brain tumors from multivariate magnetic resonance imaging and spectroscopy. Artificial Intelligence in Medicine. 2008;43(May):61–74
  40. Duda RO, Hart PE, Stork DG. Pattern classification. New York: Wiley; 2001;
  41. Jacobs RA, Jordan MI, Nowlan SJ, Hinton GE. Adaptive mixtures of local experts. Neural Computation. 1991;3:79–87
  42. Rogova G. Combining the results of several neural network classifiers. Neural Networks. 1994;7:777–781
  43. Polikar R. Ensemble based systems in decision making. IEEE Circuits and Systems Magazine. 2006;6:21–45
  44. Li Q, Doi K. Reduction of bias and variance for evaluation of computer-aided diagnostic schemes. Medical Physics. 2006;33(April):868–875
  45. Metz CE, Herman BA, Shen JH. Maximum likelihood estimation of receiver operating characteristic (ROC) curves from continuously-distributed data. Statistics in Medicine. 1998;17(May):1033–1053
  46. Mougiakakou SG, Valavanis IK, Nikita A, Nikita KS. Differential diagnosis of CT focal liver lesions using texture features, feature selection and ensemble driven classifiers. Artificial Intelligence in Medicine. 2007;41(September):25–37

PII: S0933-3657(10)00054-0

doi: 10.1016/j.artmed.2010.04.011

Artificial Intelligence in Medicine
Volume 50, Issue 1 , Pages 43-53 , September 2010

Article Tools

* Image Usage & Resolution