Spectral Analysis of Digital Images as a New Method in Diagnosis of Malignant Melanoma

Abstract

Hotsko Oksana, Galaychuk Igor. Spectral Analysis of Digital Images as a New Method in Diagnosis of Malignant Melanoma. Journal of Education, Health and Sport. 2015;5(10):340-349. ISSN 2391-8306. DOIhttp://dx.doi.org/10.5281/zenodo.42599

http://ojs.ukw.edu.pl/index.php/johs/article/view/2015%3B5%2810%29%3A340-349

https://pbn.nauka.gov.pl/works/686454

Formerly Journal of Health Sciences. ISSN 1429-9623 / 2300-665X. Archives 2011–2014http://journal.rsw.edu.pl/index.php/JHS/issue/archive

Deklaracja.

Specyfika i zawartość merytoryczna czasopisma nie ulega zmianie.

Zgodnie z informacją MNiSW z dnia 2 czerwca 2014 r., że w roku 2014 nie będzie przeprowadzana ocena czasopism naukowych; czasopismo o zmienionym tytule otrzymuje tyle samo punktów co na wykazie czasopism naukowych z dnia 31 grudnia 2014 r.

The journal has had 5 points in Ministry of Science and Higher Education of Poland parametric evaluation. Part B item 1089. (31.12.2014).

© The Author (s) 2015;

This article is published with open access at Licensee Open Journal Systems of Kazimierz Wielki University in Bydgoszcz, Poland and Radom University in Radom, Poland

Open Access. This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium,

provided the original author(s) and source are credited. This is an open access article licensed under the terms of the Creative Commons Attribution Non Commercial License

(http://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted, non commercial use, distribution and reproduction in any medium, provided the work is properly cited.

This is an open access article licensed under the terms of the Creative Commons Attribution Non Commercial License (http://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted, non commercial

use, distribution and reproduction in any medium, provided the work is properly cited.

The authors declare that there is no conflict of interests regarding the publication of this paper.

Received: 25.09.2015. Revised 25.10.2015. Accepted: 31.10.2015.

Spectral Analysis of Digital Images as a New Method in Diagnosis of Malignant Melanoma

Oksana Hotsko, Igor Galaychuk

I.Ya. Horbachevsky Ternopil State Medical University,

Ternopil, Ukraine

            Summary

                Spectrophotometric analysis (SPM) of the digital images of pigmented skin tumours was performed using ImageJ software. The following algorithm: selection of the area and segmentation of the tumours nevus image; decomposition of the image into primary colours (red, green and blue), and extraction of their histograms; histographic evaluation of the colour spectra using absolute and relative values and charts built using Microsoft Excel. To validate the method, the SPM values were correlated with the results of morphological study of the excised nevi.

We determined that the red color spectrogram is the main diagnostic measure. In it, we selected the following parameters: intensity of the red colour (I_start, I_max,) and АUC₁₂₈ the total area under the graph.

The cluster analysis of the spectrophotograms to determine thresholds of the intensity of the red colour needed to differentiate malignant melanoma and dysplastic nevi were done. Melanomas can be characterized using the following values: I_start ≤ 18.5 units, I_max ≤ 115.5 units, AUC₁₂₈ ≥ 5442.1. Opposite values are characteristic for dysplastic nevi.

Keywords: skin melanoma, melanocytic nevi, spectrophotometric analysis.