Paper
7 December 2017 Rosacea assessment by erythema index and principal component analysis segmentation maps
Author Affiliations +
Proceedings Volume 10592, Biophotonics—Riga 2017; 1059204 (2017) https://doi.org/10.1117/12.2285175
Event: The Second International Conference "Biophotonics-Riga 2017", 2017, Riga, Latvia
Abstract
RGB images of rosacea were analyzed using segmentation maps of principal component analysis (PCA) and erythema index (EI). Areas of segmented clusters were compared to Clinician’s Erythema Assessment (CEA) values given by two dermatologists. The results show that visible blood vessels are segmented more precisely on maps of the erythema index and the third principal component (PC3). In many cases, a distribution of clusters on EI and PC3 maps are very similar. Mean values of clusters’ areas on these maps show a decrease of the area of blood vessels and erythema and an increase of lighter skin area after the therapy for the patients with diagnosis CEA = 2 on the first visit and CEA=1 on the second visit. This study shows that EI and PC3 maps are more useful than the maps of the first (PC1) and second (PC2) principal components for indicating vascular structures and erythema on the skin of rosacea patients and therapy monitoring.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Ilona Kuzmina, Uldis Rubins, Inga Saknite, and Janis Spigulis "Rosacea assessment by erythema index and principal component analysis segmentation maps ", Proc. SPIE 10592, Biophotonics—Riga 2017, 1059204 (7 December 2017); https://doi.org/10.1117/12.2285175
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Image segmentation

Skin

RGB color model

Blood vessels

Principal component analysis

Image analysis

Image processing

Back to Top