Melanoma recognition using representative and discriminative kernel classifiers
2006 (English)In: Computer Vision Approaches To Medical Image Analysis / [ed] Beichel, RR, 2006, Vol. 4241, 1-12 p.Conference paper (Refereed)
Malignant melanoma is the most deadly form of skin lesion. Early diagnosis is of critical importance to patient survival. Existent visual recognition algorithms for skin lesions classification focus mostly on segmentation and feature extraction. In this paper instead we put the emphasis on the learning process by using two kernel-based classifiers. We chose a discriminative approach using support vector machines, and a probabilistic approach using spin glass-Markov random fields. We benchmarked these algorithms against the (to our knowledge) state-of-the-art method on melanoma recognition, exploring how performance changes by using color or textural features, and how it is affected by the quality of the segmentation mask. We show with extensive experiments that the support vector machine approach outperforms the existing method and, on two classes out of three, it achieves performances comparable to those obtained by expert clinicians.
Place, publisher, year, edition, pages
2006. Vol. 4241, 1-12 p.
, Lecture Notes in Computer Science, ISSN 0302-9743 ; 4241
Computer and Information Science
IdentifiersURN: urn:nbn:se:kth:diva-42003ISI: 000241602500001ScopusID: 2-s2.0-33750738872ISBN: 3-540-46257-0OAI: oai:DiVA.org:kth-42003DiVA: diva2:445821
2nd International ECCV Workshop on Computer Vision Approaches to Medical Image Analysis, CVAMIA 2006; Graz; 12 May 2006 through 12 May 2006
QC 201110052011-10-052011-10-052011-10-05Bibliographically approved