Premium
Automated diagnosis of pigmented skin lesions
Author(s) -
Rubegni Pietro,
Cevenini Gabriele,
Burroni Marco,
Perotti Roberto,
Dell'Eva Giordana,
Sbano Paolo,
Miracco Clelia,
Luzi Pietro,
Tosi Piero,
Barbini Paolo,
Andreassi Lucio
Publication year - 2002
Publication title -
international journal of cancer
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.475
H-Index - 234
eISSN - 1097-0215
pISSN - 0020-7136
DOI - 10.1002/ijc.10620
Subject(s) - artificial neural network , artificial intelligence , receiver operating characteristic , pattern recognition (psychology) , diagnostic accuracy , computer science , spectrum analyzer , skin lesion , melanoma , medical diagnosis , feature (linguistics) , medicine , pathology , radiology , machine learning , telecommunications , linguistics , philosophy , cancer research
Since advanced melanoma remains practically incurable, early detection is an important step toward a reduction in mortality. High expectations are entertained for a technique known as dermoscopy or epiluminescence light microscopy; however, evaluation of pigmented skin lesions by this method is often extremely complex and subjective. To obviate the problem of qualitative interpretation, methods based on mathematical analysis of pigmented skin lesions, such as digital dermoscopy analysis, have been developed. In the present study, we used a digital dermoscopy analyzer (DBDermo‐Mips system) to evaluate a series of 588 excised, clinically atypical, flat pigmented skin lesions (371 benign, 217 malignant). The analyzer evaluated 48 parameters grouped into 4 categories (geometries, colors, textures and islands of color), which were used to train an artificial neural network. To evaluate the diagnostic performance of the neural network and to check it during the training process, we used the error area over the receiver operating characteristic curve. The discriminating power of the digital dermoscopy analyzer plus artificial neural network was compared with histologic diagnosis. A feature selection procedure indicated that as few as 13 of the variables were sufficient to discriminate the 2 groups of lesions, and this also ensured high generalization power. The artificial neural network designed with these variables enabled a diagnostic accuracy of about 94%. In conclusion, the good diagnostic performance and high speed in reading and analyzing lesions (real time) of our method constitute an important step in the direction of automated diagnosis of pigmented skin lesions. © 2002 Wiley‐Liss, Inc.