Classifying Pigmented Skin Lesions with Machine Learning Methods

  • Stephan Dreiseitl
  • Harald Kittler
  • Harald Ganster
  • Michael Binder
Conference paper
Part of the Perspectives in Neural Computing book series (PERSPECT.NEURAL)


We use a data set of 1619 pigmented skin lesions images from three categories (common nevi, dysplastic nevi, and melanoma) to investigate the performance of four machine learning methods on the problem of classifying lesion images. The methods used were k-nearest neighbors, logistic regression, artificial neural networks, and support vector machines. The data sets were used to train the algorithms on the following tasks: to distinguish common nevi from dysplastic nevi and melanoma, and to distinguish melanoma from common nevi and dysplastic nevi. Receiver operating characteristic curves were used to summarize the performance of the models.

Three of the methods (logistic regression, artificial neural networks, and support vector machines) achieved very good results (area under curve values about 0.96) on the problem of distinguishing melanoma from common and dysplastic nevi, and good results (area under curve values about 0.82) for the problem of distinguishing common nevi from dysplastic nevi and melanoma. The performance of k-nearest neighbors models was about 3 percentage points worse than that of the other methods. These results show that for classification problems in the domain of pigmented skin lesions, excellent results can be achieved with several algorithms.


Support Vector Machine Logistic Regression Artificial Neural Network Radial Basis Function Kernel Dysplastic Nevus 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    C.M. Grin, A.W. Kopf, B. Welkovich, R.S. Bart, and M.J. Levenstein. Accuracy in the clinical diagnosis of malignant melanoma. Arch of Dermat, 126:763–766, 1990.CrossRefGoogle Scholar
  2. [2]
    V.N. Vapnik. The Nature of Statistical Learning Theory. Springer Verlag, 1995.Google Scholar
  3. [3]
    V.N. Vapnik. Statistical Learning Theory. John Wiley & Sons, 1998.Google Scholar
  4. [4]
    J.A. Hanley and B.J. McNeil. The meaning and use of the area under the receiver operating characteristic (ROC) curve. Radiol, 143:29–36, 1982.Google Scholar
  5. [5]
    M. Binder, M. Schwarz, and A. Winkler et al. Epiluminescence microscopy. A useful tool for the diagnosis of pigmented skin lesions for formally trained dermatologists. Arch of Dermal, 131:286–291, 1995.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag London 2000

Authors and Affiliations

  • Stephan Dreiseitl
    • 1
  • Harald Kittler
    • 2
  • Harald Ganster
    • 3
  • Michael Binder
    • 4
  1. 1.Dept. of Software Engineering for MedicinePolytechnic University of Upper AustriaHagenbergAustria
  2. 2.Department of DermatologyUniversity of Vienna Medical SchoolViennaAustria
  3. 3.Institute for Computer Graphics and VisionTechnical UniversityGrazAustria
  4. 4.Decision Systems GroupBrigham and Women’s HospitalBostonUSA

Personalised recommendations