Model for Estimating the Heterogeneity of the Distribution of Globule Characteristics in Images of Skin Neoplasms

Autor: Vasily Yu. Sergeev, O. B. Tamrazova, A. N. Pronichev, V. S. Kozlov, Alina O. Lim, V. G. Nikitaev, Vladimir Yu. Selchuk
Rok vydání: 2021
Předmět:
Zdroj: Measurement Techniques. 64:778-783
ISSN: 1573-8906
0543-1972
DOI: 10.1007/s11018-022-02003-w
Popis: The problem of skin melanoma diagnostics from digital images of the tumor is considered. Clinical algorithms for detecting skin melanoma are briefly described. An overview of the works devoted to the automated assessment of the asymmetry of the distribution of shape, color, area of globules – important signs of melanoma – is given. A model for estimating the heterogeneity of the distribution of the characteristics of globules on digital images in the skin neoplasms diagnosis is developed and models of signs of heterogeneity of this distribution are proposed. The comparative evaluation of the proposed models was carried out experimentally using a software system developed in C++. The most informative features are identified. The greatest accuracy 93 % in estimating the heterogeneity of the distribution of the characteristics of globules was shown by the sign “the reduced inverse of the greatest frequency of occurrence of the measured areas of globules”. The results obtained can be applied in the development of systems to support medical decision-making in the diagnosis of melanoma.
Databáze: OpenAIRE