DERMOSCOPY Evaluation of a Program for the Automatic Dermoscopic Diagnosis of Melanoma in a General Dermatology Setting ALEJANDRO FUEYO-CASADO, MD, FRANCISCO VA ´ ZQUEZ-LO ´ PEZ, MD, JESUS SANCHEZ-MARTIN, MD, BEGON ˜ A GARCIA-GARCIA, MD, AND NARCISO PE ´ REZ-OLIVA, MD The authors have indicated no significant interest with commercial supporters. T he accuracy of automated programs for the computerized, dermoscopic diagnosis of mela- noma has been found to be similar or superior to that of expert clinicians, 1,2 even in evidence-based meta- analysis. 1 Most of this research has been performed in pigmented skin lesion (PSL) units, where the number of diagnostically challenging PSLs is high. We consider it of interest to expand the investigation in a different setting. Our aim was to evaluate the usefulness of a commercially available program for the dermoscopic diagnosis of melanoma. The study was mainly focused on the detection of obvious melanoma at the first ex- amination of patients, in the daily routine practice of a general dermatology consultancy, where it has been suggested that the value of these programs would be great. 3 Patients and Methods Setting This study was conducted in a real-time, daily rou- tine practice of a general dermatology consultancy of a tertiary teaching hospital in Oviedo, northern Spain. Methods The accuracy of an automated program for diagnosis of melanoma (Tuebingen Moleanalyzer (MA), in Fotofinder dermoscope, TeachScreen Software GmbH, Bad Birnbach, Germany) was evaluated according to the basic quality requirements of Rosado and colleagues. 1 Subjects were randomly selected adult patients with melanocytic skin lesions (MSLs) examined from March to September 2007. According to the indications of the manufacturer, patients with non-MSLs were excluded. All clinical forms of MSL were evaluated, because neither the manufacturer nor the authors of the program pro- vide other restriction criteria. 4 MSLs were classified into two categories: those need- ing excision (suspected melanoma lesions) and those not requiring excision at the time of first examination (banal-appearing MSLs and MSLs to be followed up). A panel of three general dermatologists (not special- izing in PSLs) performed the clinical classification ac- cording to history of change, physical examination, and examination with a hand-held dermoscope with polarized light (Dermlite Pro 3Gen, LLC, San Juan Capistrano, CA). The Fotofinder MA provided a three-color scoring classification of MSLs: green (typ- ical MSL); yellow (somewhat atypical MSL that should be reexamined), and red (high probability of being melanoma). Because the study was designed to evaluate this program as an instrument to help decide whether MSLs should be removed at the time of the first examination, ‘‘non-red’’ scores (yellow and green) were grouped in the same category (MSLs not re- quiring excision at the time of first examination). & 2009 by the American Society for Dermatologic Surgery, Inc.  Published by Wiley Periodicals, Inc.  ISSN: 1076-0512  Dermatol Surg 2009;35:257–262  DOI: 10.1111/j.1524-4725.2008.34421.x 257 All authors are affiliated with Hospital Universitario Central de Asturias, University of Oviedo, Oviedo, Spain