Методы машинного обучения в радиомике для анализа кардиоваскулярных изображений

Abstract

This article outlines the idea of applying machine learning techniques in nuclear medicine to analyze cardiovascular images. The work includes segmentation of the affected area of the heart on the polar map, isolation of radiomic indicators from the ROI, and classification of patients into healthy individuals and patients with abnormalities. The article also presents some of the results of the initial study and the first stage of work.