Comparative Analysis of Deep Neural Network and Texture-Based Classifiers for Recognition of Acute Stroke using Non-Contrast CT Images

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

This work presents a computer technology for automatic recognition of acute stroke using non-contrast computed tomography brain images. The early diagnosis of acute stroke is of primary importance for deciding on a method for further treatment, and the developed system aims at assisting a radiology specialist in the decision making process. We consider deep neural network and texture-based classifiers in order to compare their efficiency on a moderate-sized sample of patients with acute stroke. We use U-net as a basic architecture of the neural network, and Haralick textural features, extracted from images, for kNN, SVM, Random Forest and Adaboost classifiers. Experiments with real CT images using cross-validation technique show that deep neural network outperforms the considered texture-based classifiers; however, the latter are faster in training. We demonstrate that texture-based approach is able to give potentially useful additional information for stroke recognition, such as estimates of textural features importance; visualization of differences in positive and negative class distributions.

Original languageEnglish
Title of host publicationProceedings - 2020 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages376-379
Number of pages4
ISBN (Electronic)9781728131658
DOIs
Publication statusPublished - 1 May 2020
Event2020 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2020 - Yekaterinburg, Russian Federation
Duration: 14 May 202015 May 2020

Publication series

NameProceedings - 2020 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2020

Conference

Conference2020 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2020
CountryRussian Federation
CityYekaterinburg
Period14.05.202015.05.2020

Keywords

  • acute stroke
  • classification
  • deep neural network
  • texture segmentation
  • U-net

Fingerprint Dive into the research topics of 'Comparative Analysis of Deep Neural Network and Texture-Based Classifiers for Recognition of Acute Stroke using Non-Contrast CT Images'. Together they form a unique fingerprint.

Cite this