Share Email Print

Proceedings Paper

Towards to deep neural network application with limited training data: synthesis of melanoma's diffuse reflectance spectral images
Author(s): Katrina Bolochko; Dmitrijs Bliznuks; Dilshat Uteshev; Ilze Lihacova; Alexey Lihachev; Yuriy Chizhov; Andrey Bondarenko
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

The goal of our study is to train artificial neural networks (ANN) using multispectral images of melanoma. Since the number of multispectral images of melanomas is limited, we offer to synthesize them from multispectral images of benign skin lesions. We used the previously created melanoma diagnostic criterion p'. This criterion is calculated from multispectral images of skin lesions captured under 526nm, 663nm, and 964nm LED illumination. We synthesize these three images from multispectral images of nevus so that the p' map matches the melanoma criteria (the values in the lesion area is >1, respectively). Demonstrated results show that by transforming multispectral images of benign nevus is possible to get a reliable multispectral images of melanoma usable for ANN training.

Paper Details

Date Published: 11 July 2019
PDF: 6 pages
Proc. SPIE 11074, Diffuse Optical Spectroscopy and Imaging VII, 110741O (11 July 2019); doi: 10.1117/12.2527173
Show Author Affiliations
Katrina Bolochko, Riga Technical Univ. (Latvia)
Dmitrijs Bliznuks, Riga Technical Univ. (Latvia)
Dilshat Uteshev, C.T. Co. Ltd. (Latvia)
Ilze Lihacova, Univ. of Latvia (Latvia)
Alexey Lihachev, Univ. of Latvia (Latvia)
Yuriy Chizhov, Riga Technical Univ. (Latvia)
Andrey Bondarenko, C.T. Co. Ltd. (Latvia)

Published in SPIE Proceedings Vol. 11074:
Diffuse Optical Spectroscopy and Imaging VII
Hamid Dehghani; Heidrun Wabnitz, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?