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.

[1]  Peter J. Frosch,et al.  Computerized diagnosis of skin cancer using neural networks , 1996 .

[2]  H. Kittler,et al.  Epiluminescence microscopy-based classification of pigmented skin lesions using computerized image analysis and an artificial neural network , 1998, Melanoma research.

[3]  E Claridge,et al.  Shape analysis for classification of malignant melanoma. , 1992, Journal of biomedical engineering.

[4]  Geoffrey E. Hinton,et al.  Adaptive Mixtures of Local Experts , 1991, Neural Computation.

[5]  W V Stoecker,et al.  Automatic detection of asymmetry in skin tumors. , 1992, Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society.

[6]  W V Stoecker,et al.  An automatic color segmentation algorithm with application to identification of skin tumor borders. , 1992, Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society.

[7]  A. Kopf,et al.  Early detection of malignant melanoma: The role of physician examination and self‐examination of the skin , 1985, CA: a cancer journal for clinicians.

[8]  Philippe Schmid.  Lesion detection in dermatoscopic images using anisotropic diffusion and morphological flooding , 1999, Proceedings 1999 International Conference on Image Processing (Cat. 99CH36348).

[9]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[10]  John Chung-Mong Lee,et al.  An efficient and effective texture classification approach using a new notion in wavelet theory , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[11]  R. H. Moss,et al.  Neural network diagnosis of malignant melanoma from color images , 1994, IEEE Transactions on Biomedical Engineering.

[12]  W Abmayr,et al.  Improvement of monitoring of melanocytic skin lesions with the use of a computerized acquisition and surveillance unit with a skin surface microscopic television camera. , 1996, Journal of the American Academy of Dermatology.

[13]  R. Porter,et al.  Robust rotation-invariant texture classification: wavelet, Gabor filter and GMRF based schemes , 1997 .

[14]  Clement T. Yu,et al.  Segmentation of skin cancer images , 1999, Image Vis. Comput..

[15]  Thomas Roß,et al.  Feature selection for optimized skin tumor recognition using genetic algorithms , 1999, Artif. Intell. Medicine.

[16]  Michael I. Jordan,et al.  Convergence results for the EM approach to mixtures of experts architectures , 1995, Neural Networks.

[17]  W V Stoecker,et al.  Texture in skin images: comparison of three methods to determine smoothness. , 1992, Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society.