Classification of Melanoma Thickness from Dermoscopic Images using Mobile Devices

The survival of patients with melanoma is mainly dependent on the thickness of the melanoma. The depth of the melanoma is usually give in millimeters then calculated using pathological examination following a biopsy of the suspected lesion. To avoid invasive methods in the estimation of the thickness of the melanoma prior to surgery we propose a computational image analysis system from dermoscopic images. A recent interpretable method merging logistic regression along with artificial networks (LIPU) logistic regression using initial variables and product units, which are nominal classification methods, are compared in terms of performance. In case of binary classification LIPU performs with an accuracy of 77.6% compared to all other methods while in case of the second scheme the ordinal classification methods achieve a better balance between the performances of all classes even though LIPU reports the highest overall accuracy. Citation: Dr. B. Muthusenthil, Valliammai Engineering College; Ravishanker Satish ,Valliammai Engineering College; Sivaraman. V ,Valliammai Engineering College; Srinath. R. M ,Valliammai Engineering College. "Classification of Melanoma Thickness from Dermoscopic Images using Mobile Devices." Global Research and Development Journal For Engineering 35 2018: 24 - 30.