Some Methods for Substantiating Diagnostic Decisions Made Using Machine Learning Algorithms

A G Losev; I E Popov; A Yu Petrenko; A G Gudkov; S G Vesnin; S V Chizhikov

doi:10.1007/s10527-022-10153-y

Some Methods for Substantiating Diagnostic Decisions Made Using Machine Learning Algorithms

Biomed Eng (NY). 2022;55(6):442-447. doi: 10.1007/s10527-022-10153-y. Epub 2022 Mar 7.

Authors

A G Losev¹, I E Popov¹, A Yu Petrenko¹, A G Gudkov², S G Vesnin³, S V Chizhikov^{2

4}

Affiliations

¹ Volgograd State University, Volgograd, Russia.
² Bauman Moscow State Technical University, Moscow, Russia.
³ RES Company, Moscow, Russia.
⁴ OOO NPP Tekhnologicheskie Innovatsii, Moscow, Russia.

Abstract

Various classification algorithms used in the diagnosis of breast cancer based on microwave radiometry data are considered. In particular, their principles of operation and the possibility of substantiating diagnoses using numerical data are discussed. A substantiation algorithm based on decision trees and a naive Bayesian classifier is presented. Examples of substantiation are given for breast cancer.