Computer-aided Diagnosis of Melanocytic Lesions

Background The clinical diagnosis of melanoma could be difficult for a general practitioner and in some cases for dermatologists. To enhance and support the clinical evaluation of pigmented skin lesions a computer-aided diagnosis has been introduced. Objective The purposes of this study were 1) to provide dermatologists as well as general practitioners with a computer aided device in order to produce an objective risk level of the melanocytic lesions helping the clinician during the diagnostic pathway and 2) to compare the discrimination capacity of dermatologists and general practitioners, with the system of automated analysis. Methods 477 images of melanocytic lesions (42 melanomas and 435 melanocytic) evaluated in epiluminescence microscopy and recorded with x16 magnification were selected. A training set of 22 melanomas and 218 nevi has been randomized from the dataset. The test set has been formed by the complement (the remaining 20 melanomas and 217 nevi). Furthermore a set of images consisting in 31 melanomas and 103 nevi has been selected to compare the discrimination capacity of three general practitioners and three dermatologists with experience in dermoscopy (2 years), with the system of automated analysis ADAM. Sensitivity and specificity were estimated for observer assessments and computer diagnosis.