Automated Diagnosis of Skin Cancer Using Digital Image Processing and Mixture-of-Experts
The incidence of malignant melanoma, the most lethal form of skin cancers, has risen rapidly during the last decades. Fortunately, if detected early, even malignant melanoma can be treated successfully. Thus, in recent years, methods for automated detection and diagnosis of skin cancer, particulary malignant melanoma, have elicited much interest. In this paper we present an artificial neural network approach for the classification of skin lesions. Sophisticated image processing, feature extraction, pattern recognition and methods from the field of statistics and artificial neural networks are combined in order to achieve a fast and reliable diagnosis. With this approach, for reasonably balanced training and test sets, we are able to obtain above 90% correct classification of malignant and benign skin lesions coming from the DANAOS data collection.
Unable to display preview. Download preview PDF.
- 3.R Husemann, S Tölg, W v Seelen, P Altmeyer, P J Frosch, M Stücker, K Hoffmann, and S El-Gammal. Computerised diagnosis of skin cancer using neural networks. In P Altmeyer, K Hoffmann, and M Stucker, editors, Congress on Skin Cancer and UV Radiation (1996), pages 1052–1063. Springer, 1997.Google Scholar
- 4.W Stolz, R Schiffner, L Pillet, T Vogt, H Harms, T Schindewolf, M Landthaler, and W Abmayr. Improvement of monitoring of melanocytic skin lesions with the use of a computarized acquisition and surveillance unit with a skin surface microscopic television camera. J Am Acad Dermatol, 35, 1996.Google Scholar
- 6.G Pott, R Husemann, L Eckert, T Grünendick, S Lux, and P Altmeyer. Danaos - Automated skin cancer diagnosis with neural networks. In Proc 7th EADV, 1998.Google Scholar
- 10.P Schmid. Lesion detection in dermatoscopic images using anisotropic diffusion and morphological flooding. In Proc Int Conf Image Processing (ICIP’99), volume 3, pages 449–453. IEEE Signal Process Soc, 1999.Google Scholar
- 13.J-F Liu and J C-M Lee. An efficient and effective texture classification approach using a new notion in wavelet theory. In Proc ICPR’96, pages 820–824, 1996.Google Scholar
- 14.R Porter and N Canagarajah. Robust rotation-invariant texture classification: Wavelet, Gabor filter and GMRF based schemes. IEEE Proceedings - Vision, Image and Signal Processing, 144:188ff, 1997.Google Scholar