Melanoma Recognition Using Representative and Discriminative Kernel Classifiers

Abstract

Abstract. 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.

Design and Implementation of CNN for detecting Melanoma through Image Processing

“…A lot of efforts have been dedicated in solving this challenging problem. Early investigations are achieved to apply low-level hand-crafted features to distinguish melanomas from non-melanoma skin lesions, including, color [5],shape [6] and texture [7], [8]. Some of the researchers are undertaken to employ feature selection algorithms to select proper features and utilized combinations of these low-level features to improve the recognition performance [9], [10].…”

A new Mercer sigmoid kernel for clinical data classification

“…The strong performance of Support Vector Machines (SVM) and kernel methods make them a mainstay as one of the state-of-the-art techniques for classification [1], [2], including applications to clinical research, diagnosis and prognosis [3], [4], [5]. One of the key issues in specifying an SVM solution is choosing the right kernel for the data and task, since a wrong choice can have a detrimental and possibly profound impact on classification accuracy [1], [2], [6].…”

Pattern analysis of dermoscopic images based on Markov random fields

“…Normally, the papers present one approach to cover one or some of the "letters" of the rule, that is, some are based on detecting asymmetry [7,8], borders [9][10][11][12], colour [13][14][15] or diameter [14]. There are some papers that cover the whole ABCD criterion.…”

“…But, in any case, their purpose is to separate the lesion from the healthy skin, which is not our aim (we perform a classification of the different patterns that a lesion can present). In [14] the authors present a melanoma recognition system, but the use of MRF is not for the segmentation or characterization of the coloured patterns, but as a classifier (spin glass-MRF). Their inputs are features characterizing either the C letter of the ABCD rule or the D letter of the same rule.…”