|
Algorithm for analyzing the characteristics of the pigment network in the diagnosis of melanoma
V. G. Nikitaeva, O. B. Tamrazovab, A. N. Pronicheva, V. Yu. Sergeevc, E. A. Druzhininaa a National Research Nuclear University MEPhI (Moscow Engineering Physics Institute)
b Peoples' Friendship University of Russia
c Central State Medical Academy of the Presidential Administration of Russian Federation
Abstract:
The paper offers an algorithm for analyzing the characteristics of the pigment network of
skin neoplasms. The algorithm is based on the study of the coefficient of variation of average lengths of the segments of pigment network in local areas of tumors from the average value of the lengths of the line segments pigment network throughout the area of the
tumor. The algorithm allows you to distinguish a typical pigment network from an atypical one. An atypical pigment network is an essential feature in determining early-stage
melanoma. The algorithm can be used in automated systems for supporting medical decision — making in the diagnosis of skin neoplasms.
Keywords:
digital image processing, segmentation, skeletonization, atypical pigment
network, dermatoscopy, diagnosis of melanoma.
Received: 13.10.2020 Revised: 13.10.2020 Accepted: 30.11.2020
Citation:
V. G. Nikitaev, O. B. Tamrazova, A. N. Pronichev, V. Yu. Sergeev, E. A. Druzhinina, “Algorithm for analyzing the characteristics of the pigment network in the diagnosis of melanoma”, Matem. Mod., 33:2 (2021), 67–81; Math. Models Comput. Simul., 13:5 (2021), 861–869
Linking options:
https://www.mathnet.ru/eng/mm4262 https://www.mathnet.ru/eng/mm/v33/i2/p67
|
Statistics & downloads: |
Abstract page: | 383 | Full-text PDF : | 157 | References: | 52 | First page: | 13 |
|