Logo image
Back
COVID TV-Unet: Segmenting COVID-19 chest CT images using connectivity imposed Unet
Journal article   Open access   Peer reviewed

COVID TV-Unet: Segmenting COVID-19 chest CT images using connectivity imposed Unet

Narges Saeedizadeh, Shervin Minaee, Rahele Kafieh, Shakib Yazdani and Milan Sonka
Computer Methods and Programs in Biomedicine Update, Vol.1, pp.100007-100007
2021
DOI: 10.1016/j.cmpbup.2021.100007
PMCID: PMC8056883
PMID: 34337587
url
https://doi.org/10.1016/j.cmpbup.2021.100007View
Published (Version of record) Open Access

Abstract

•A segmentation framework to detect infected chest regions in CT images.•A regularization term based on 2D-anisotropic total-variation is added to the loss function.•A relatively largescale CT segmentation dataset of around 900 images.•Identifying infected regions with mIoU rate of 99%, and a Dice score of 86%. The novel corona-virus disease (COVID-19) pandemic has caused a major outbreak in more than 200 countries around the world, leading to a severe impact on the health and life of many people globally. By October 2020, more than 44 million people were infected, and more than 1,000,000 deaths were reported. Computed Tomography (CT) images can be used as an alternative to the time-consuming RT-PCR test, to detect COVID-19. In this work we propose a segmentation framework to detect chest regions in CT images, which are infected by COVID-19. An architecture similar to a Unet model was employed to detect ground glass regions on a voxel level. As the infected regions tend to form connected components (rather than randomly distributed voxels), a suitable regularization term based on 2D-anisotropic total-variation was developed and added to the loss function. The proposed model is therefore called ”TV-Unet”. Experimental results obtained on a relatively large-scale CT segmentation dataset of around 900 images, incorporating this new regularization term leads to a 2% gain on overall segmentation performance compared to the Unet trained from scratch. Our experimental analysis, ranging from visual evaluation of the predicted segmentation results to quantitative assessment of segmentation performance (precision, recall, Dice score, and mIoU) demonstrated great ability to identify COVID-19 associated regions of the lungs, achieving a mIoU rate of over 99%, and a Dice score of around 86%.
 Computed tomography Convolutional encoder decoder COVID-19 Deep learning Image segmentation Total variation

Details

Metrics

89 Record Views
Logo image