TY - GEN
T1 - Challenges of automatic processing of large amount of skin lesion multispectral data
AU - Lihacova, I.
AU - Cibulska, E.
AU - Lihachev, A.
AU - Lange, M.
AU - Plorina, E. V.
AU - Bliznuks, D.
AU - Derjabo, A.
AU - Kiss, N.
N1 - Funding Information:
This work has been supported by European Regional Development Fund projects “Development and clinical validation of a novel cost effective multi-modal methodology for early diagnostics of skin cancers” (No. 1.1.1.2/16/I/001, agreement No. 1.1.1.2/VIAA/1/16/052) and “Time-resolved autofluorescence methodology for non-invasive skin cancer diagnostics” (No. 1.1.1.2/16/I/001, agreement No. 1.1.1.2/VIAA/1/16/014).
Publisher Copyright:
© 2020 SPIE.
Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.
PY - 2020
Y1 - 2020
N2 - This work will describe the challenges involved in setting up automatic processing for a large differentiated data set. In this study, a multispectral (skin diffuse reflection images using 526nm (green), 663nm (red), and 964nm (infrared) illumination and autofluorescence (AF) image using 405 nm excitation) data set with 756 lesions (3024 images) was processed. Previously, using MATLAB software, finding markers, correctly segmenting images with dark edges and image alignment were the main causes of the problems in automatic data processing. To improve automatic processing and eliminate the use of licensed software, the latter was substituted with the open source Python environment. For more precise segmentation of skin markers and skin lesions, as well for image alignment, the processing of artificial neural networks was utilized. The resulting processing method solves most of the issues of the MATLAB script. However, for even more accurate results, it is necessary to provide more accurate ground-truth segmentation masks and generate more input data to increase the training image database by using data augmentation.
AB - This work will describe the challenges involved in setting up automatic processing for a large differentiated data set. In this study, a multispectral (skin diffuse reflection images using 526nm (green), 663nm (red), and 964nm (infrared) illumination and autofluorescence (AF) image using 405 nm excitation) data set with 756 lesions (3024 images) was processed. Previously, using MATLAB software, finding markers, correctly segmenting images with dark edges and image alignment were the main causes of the problems in automatic data processing. To improve automatic processing and eliminate the use of licensed software, the latter was substituted with the open source Python environment. For more precise segmentation of skin markers and skin lesions, as well for image alignment, the processing of artificial neural networks was utilized. The resulting processing method solves most of the issues of the MATLAB script. However, for even more accurate results, it is necessary to provide more accurate ground-truth segmentation masks and generate more input data to increase the training image database by using data augmentation.
KW - artificial neural networks
KW - multispectral melanoma diagnostics
KW - skin lesion segmentation
UR - http://www.scopus.com/inward/record.url?scp=85096538904&partnerID=8YFLogxK
U2 - 10.1117/12.2582049
DO - 10.1117/12.2582049
M3 - Conference contribution
AN - SCOPUS:85096538904
T3 - Proceedings of SPIE - The International Society for Optical Engineering
BT - Biophotonics - Riga 2020
A2 - Spigulis, Janis
PB - SPIE
T2 - Biophotonics - Riga 2020
Y2 - 24 August 2020 through 25 August 2020
ER -