Detection and severity classification of COVID-19 in CT images using deep learning
Author | Qiblawey Y. |
Author | Tahir A. |
Author | Chowdhury M.E.H. |
Author | Khandakar A. |
Author | Kiranyaz, Mustafa Serkan |
Author | Rahman T. |
Author | Ibtehaz N. |
Author | Mahmud S. |
Author | Al Maadeed S. |
Author | Musharavati F. |
Author | Ayari M.A. |
Available date | 2022-04-26T12:31:19Z |
Publication Date | 2021 |
Publication Name | Diagnostics |
Resource | Scopus |
Identifier | http://dx.doi.org/10.3390/diagnostics11050893 |
Abstract | Detecting COVID-19 at an early stage is essential to reduce the mortality risk of the patients. In this study, a cascaded system is proposed to segment the lung, detect, localize, and quantify COVID-19 infections from computed tomography images. An extensive set of experiments were performed using Encoder?Decoder Convolutional Neural Networks (ED-CNNs), UNet, and Feature Pyramid Network (FPN), with different backbone (encoder) structures using the variants of DenseNet and ResNet. The conducted experiments for lung region segmentation showed a Dice Similarity Coefficient (DSC) of 97.19% and Intersection over Union (IoU) of 95.10% using U-Net model with the DenseNet 161 encoder. Furthermore, the proposed system achieved an elegant performance for COVID-19 infection segmentation with a DSC of 94.13% and IoU of 91.85% using the FPN with DenseNet201 encoder. The proposed system can reliably localize infections of various shapes and sizes, especially small infection regions, which are rarely considered in recent studies. Moreover, the proposed system achieved high COVID-19 detection performance with 99.64% sensitivity and 98.72% specificity. Finally, the system was able to discriminate between different severity levels of COVID-19 infection over a dataset of 1110 subjects with sensitivity values of 98.3%, 71.2%, 77.8%, and 100% for mild, moderate, severe, and critical, respectively. |
Language | en |
Publisher | MDPI |
Subject | Article computer assisted tomography controlled study convolutional neural network coronavirus disease 2019 deep learning diagnostic accuracy diagnostic test accuracy study disease severity human image segmentation lung volume major clinical study sensitivity and specificity |
Type | Article |
Issue Number | 5 |
Volume Number | 11 |
Files in this item
Files | Size | Format | View |
---|---|---|---|
There are no files associated with this item. |
This item appears in the following Collection(s)
-
Computer Science & Engineering [2284 items ]
-
COVID-19 Research [802 items ]
-
Electrical Engineering [2611 items ]