Article ID Journal Published Year Pages File Type
3204588 Journal of the American Academy of Dermatology 2015 8 Pages PDF
Abstract

BackgroundComputer-assisted diagnosis of dermoscopic images of skin lesions has the potential to improve melanoma early detection.ObjectiveWe sought to evaluate the performance of a novel classifier that uses decision forest classification of dermoscopic images to generate a lesion severity score.MethodsSeverity scores were calculated for 173 dermoscopic images of skin lesions with known histologic diagnosis (39 melanomas, 14 nonmelanoma skin cancers, and 120 benign lesions). A threshold score was used to measure classifier sensitivity and specificity. A reader study was conducted to compare the sensitivity and specificity of the classifier with those of 30 dermatology clinicians.ResultsThe classifier sensitivity for melanoma was 97.4%; specificity was 44.2% in a test set of images. In the reader study, the classifier's sensitivity to melanoma was higher (P < .001) and specificity was lower (P < .001) than that of clinicians.LimitationsThis is a retrospective study using existing images primarily chosen for biopsy by a dermatologist. The size of the test set is small.ConclusionsOur classifier may aid clinicians in deciding if a skin lesion should be biopsied and can easily be incorporated into a portable tool (that uses no proprietary equipment) that could aid clinicians in noninvasively evaluating cutaneous lesions.

Related Topics
Health Sciences Medicine and Dentistry Dermatology
Authors
, , , , , , ,