Hyperspecral skin imaging with artificial neural networks validated by optical biotissue phantoms

A. Bykov*, E. Zherebtsov, V. Dremin, A. Popov, A. Doronin, I. Meglinski

*Corresponding author for this work

Research output: Chapter in Book/Published conference outputConference publication

Abstract

State-of-the-art micro-optic multichannel matrix sensor combined with the tunable Fabry-Perot micro interferometer enables a compact diagnostic device sensitive to the changes of the oxygen saturation as well as the blood volume fraction of human skin. The possibility of using Monte-Carlo modelling for neural network training in the problem of hyperspectral image processing has been demonstrated and validated using biotissue phantom and human skin in vivo. The proposed approach enables a tool combining both the speed of neural network processing and the accuracy and flexibility of Monte-Carlo modelling.

Original languageEnglish
Title of host publicationComputational Optical Sensing and Imaging, COSI 2019
ISBN (Electronic)9781557528209
DOIs
Publication statusPublished - 27 Jun 2019
EventComputational Optical Sensing and Imaging, COSI 2019 - Munich, Germany
Duration: 24 Jun 201927 Jun 2019

Conference

ConferenceComputational Optical Sensing and Imaging, COSI 2019
CountryGermany
CityMunich
Period24/06/1927/06/19

Bibliographical note

Funding Information:
Authors acknowledge the support of the Academy of Finland (grants: 290596, 296408).

Publisher Copyright:
© OSA 2019 © 2019 The Author(s)

Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.

Fingerprint

Dive into the research topics of 'Hyperspecral skin imaging with artificial neural networks validated by optical biotissue phantoms'. Together they form a unique fingerprint.

Cite this