Open AccessModification of a melanoma discrimination index derived from hyperspectral data: a clinical trial conducted in 2 centers between March 2011 and December 2013
Author(s) -
Nagaoka T.,
Kiyohara Y.,
Koga H.,
Nakamura A.,
Saida T.,
Sota T.
Publication year - 2015
Publication title -
skin research and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.521
H-Index - 69
eISSN - 1600-0846
pISSN - 0909-752X
DOI - 10.1111/srt.12188
Subject(s) - hyperspectral imaging , medicine , index (typography) , melanoma , dermatology , sensitivity (control systems) , artificial intelligence , computer science , cancer research , world wide web , electronic engineering , engineering
Background The morphology of pigmented skin lesions ( PSL s) is predominantly a result of varying concentrations and distributions of pigmented molecules such as melanin and hemoglobin. Based on these differences and the fact that their information is contained in cutaneous spectra, a hyperspectral imager ( HSI ) for pigmented melanoma and a single discrimination index derived from the resultant hyperspectral data are proposed. Objective To develop and evaluate a new discrimination index for melanomas, compared to the previous index. Methods A HSI , which is convenient for both patients and clinicians, was newly developed and used in a clinical trial conducted in 2 centers with 80 patients with primary lesions and 17 volunteers between March 2011 and December 2013. There were 24 melanomas and 110 other PSL s. A previously proposed discrimination index was used without modifications. A new index, which emphasized the essential features of melanoma, was proposed, and its performance was examined. For each index, a threshold value was set to minimize the average value of the false positive and false negative fractions. The performances of both indices were compared. Results The sensitivity and specificity of the old index were 75% and 97%, respectively, while those of the new index were 96% and 87%. Conclusion The new index had a higher sensitivity and adequate specificity, indicating that it is more useful than the old index.