An Automated System for Classification of Radiographs of the Breast.

The article discusses the basic principles of design and testing of an automated system for classification of radiographs of the breast. The principle of operation of the automated system is based on the segmentation of radiographs into rectangular areas with their subsequent deformation according t...

Full description

Bibliographic Details
Published in:Biomedical Engineering Vol. 53; no. 6; pp. 425 - 429
Main Authors: Dabagov, A. R., Gorbunov, V. A., Filist, S. A., Malyutina, I. A., Kondrashov, D. S.
Format: diagnostic images tables/charts Journal Article
Published: Springer Nature Mar2020
Online Access:View this record in EBSCOhost
Description
Summary:The article discusses the basic principles of design and testing of an automated system for classification of radiographs of the breast. The principle of operation of the automated system is based on the segmentation of radiographs into rectangular areas with their subsequent deformation according to threshold-type criteria of homogeneity. Classification of the selected segments is carried out by neural network classifiers operating in the space of information-bearing features that is built using multi-method algorithms. The software is implemented in MATLAB 2018b. The quality of classification using the developed software was tested on control samples.