Abstract | ||
---|---|---|
Melanoma, one of the most aggressive types of cancer, can be healed, if recognized in early stages. In order to automate the early recognition of skin cancer, a system that analyses digital epiluminescence microscopic images is used. After segmentation, 33 features representing shape and radiometric properties are calculated. In this paper the quality of the features is evaluated by applying several feature selection methods. The results show that with each selection method the feature set can be reduced to dimension four with nearly no loss of information. Results with classification rates of up to 75% are achieved and realtions between selected features and medical criteria are observed. |
Year | DOI | Venue |
---|---|---|
1998 | 10.1109/ICPR.1998.712040 | ICPR |
Keywords | Field | DocType |
classification rate,selection method,feature selection,feature set,feature selection method,skin cancer,early stage,selected feature,aggressive type,digital epiluminescence microscopic image,early recognition,melanoma recognition,image segmentation,computer graphics,optical microscopy,cancer,microscopy,image classification,shape,image analysis,radiometry,bioluminescence,skin,image recognition,biomedical imaging | Computer vision,Pattern recognition,Feature selection,Medical imaging,Segmentation,Computer science,Skin cancer,Image segmentation,Feature set,Artificial intelligence,Melanoma,Contextual image classification | Conference |
ISSN | ISBN | Citations |
1051-4651 | 0-8186-8512-3-2 | 3 |
PageRank | References | Authors |
0.51 | 2 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
r rohrer | 1 | 3 | 0.51 |
Harald Ganster | 2 | 169 | 11.82 |
axel pinz | 3 | 8 | 1.02 |
michael binder | 4 | 3 | 0.51 |