Open AccessHyperspectral in vivo analysis of normal skin chromophores and visualization of oncological pathologies
Author(s) -
Violetta P. Sherendak,
Ivan А. Bratchenko,
Oleg O. Myakinin,
Pavel N. Volkhin,
Yulia А. Khristoforova,
Alexander A. Moryatov,
А. С. Мачихин,
В. Э. Пожар,
S Kozlov,
Valery P. Zakharov
Publication year - 2019
Publication title -
kompʹûternaâ optika
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.491
H-Index - 29
eISSN - 2412-6179
pISSN - 0134-2452
DOI - 10.18287/2412-6179-2019-43-4-661-670
Subject(s) - hyperspectral imaging , artificial intelligence , visualization , computer science , computer vision , set (abstract data type) , pattern recognition (psychology) , programming language
In the paper, we present test results of methods for the noninvasive diagnosis of skin neoplasms, based on the hyperspectral registration of images by using a camera with an acousto-optic tunable filter. For the identification of oncological pathologies, an integral spectral index has been proposed for a set of concentric regions around the source of neoplasm growth for the tissue sample under study. As well as taking account of changes in the spectral properties of the tissue, the introduced index indirectly takes into account classical ABCD dermatoscopic features: asymmetry, border irregularity, color diversity, and the tumor diameter. Results of training set separating are presented and the applicability of the proposed approaches to the clinical practice is shown.