A Deep Learning Model for Segmentation of COVID-19 Infections Using CT scans

Описание

Тип публикации: доклад, тезисы доклада, статья из сборника материалов конференций

Конференция: 1st International Conference on Advanced Research in Pure and Applied Science, ICARPAS 2021

Год издания: 2022

Идентификатор DOI: 10.1063/5.0093739

Ключевые слова: color-map, computed tomography, covid-19, deep learning, image segmentation

Аннотация: Computed tomography is critical in diagnosing and assessing COVID-19 infection. Coronavirus (COVID-19) spread around the world in 2020, leaving the world facing an acute health crisis. The automatic deletion of lung infection on computed tomography scan (CT) images offers great potential for improving traditional healthcare strategПоказать полностьюies for treating COVID-19. However, the detection of lesions via CT imaging faces many challenges, including high contrast in infection characteristics and low contrast intensity between infection and normal tissues. Early diagnosis is an effective way to treat this condition. Then offered a deep learning pipeline consists of three different deep learning structures for generating and segmenting computed tomography of lungs and COVID-19 infection. In addition to this image pre-processing, image magnification and parameter correction based on the color model and model similarity were used to improve the diagnostic process (medium and strong infection areas). © 2022 American Institute of Physics Inc.. All rights reserved.

Ссылки на полный текст

Издание

Журнал: AIP Conference Proceedings

Выпуск журнала: Vol. 2398

Номера страниц: 50005

ISSN журнала: 0094243X

Издатель: American Institute of Physics Inc.

Персоны

  • Hamad Y.A. (Deep Learning Laboratory, Siberian Federal University, Krasnoyarsk, Russian Federation, Collage of Information Technology, Imam Jafar Al-Sadiq University, Kirkuk, Iraq)
  • Kadum J. (Department of Computer Science, College of Science, University of Diyala, Diyala, Iraq)
  • Rashid A.A. (Collage of Information Technology, Imam Jafar Al-Sadiq University, Kirkuk, Iraq)
  • Mohsen A.H. (Department of Computer Science, Al-Qalam University College, Kirkuk, Iraq)
  • Safonova A. (Deep Learning Laboratory, Siberian Federal University, Krasnoyarsk, Russian Federation)

Вхождение в базы данных