(Department of Computer Science, City, University of London, Northampton Square, London, United Kingdom, United Kingdom)*corresponding author
AbstractThis paper compares the models for the detection and segmentation of Ground Glass Opacity and Consolidation in chest CT scans. These lesion areas are often associated both with common pneumonia and COVID-19. We train a Mask R-CNN model to segment these areas with high accuracy using three approaches: merging masks for these lesions into one, deleting the mask for Consolidation, and using both masks separately. The best model achieves the mean average precision of 44.68% using MS COCO criterion on the segmentation across all accuracy thresholds. The classification model, COVID-CT-Mask-Net, learns to predict the presence of COVID-19 vs. common pneumonia vs. control. The model achieves the 93.88% COVID-19 sensitivity, 95.64% overall accuracy, 95.06% common pneumonia sensitivity, and 96.91% true-negative rate on the COVIDx-CT test split (21192 CT scans) using a small fraction of the training data. We also analyze the effect of the Non-Maximum Suppression of overlapping object predictions, both on the segmentation and classification accuracy. The full source code, models, and pre-trained weights are available on https://github.com/AlexTS1980/COVID-CT-Mask-Net.
KeywordsCOVID-19; Lesion Segmentation; Pneumonia Classification; Mask R-CNN
|
DOIhttps://doi.org/10.31763/sitech.v1i2.202 |
Article metrics10.31763/sitech.v1i2.202 Abstract views : 2179 | PDF views : 902 |
Cite |
Full Text Download
|
References
T. Yan, P. K. Wong, H. Ren, H. Wang, J. Wang, and Y. Li, “Automatic distinction between COVID-19 and common pneumonia using multi-scale convolutional neural network on chest CT scans,” Chaos, Solitons & Fractals, vol. 140, p. 110153, Nov. 2020, doi: 10.1016/j.chaos.2020.110153.
X. Li, X. Fang, Y. Bian, and J. Lu, “Comparison of chest CT findings between COVID-19 pneumonia and other types of viral pneumonia: a two-center retrospective study,” Eur. Radiol., vol. 30, no. 10, pp. 5470–5478, Oct. 2020, doi: 10.1007/s00330-020-06925-3.
W. Zhao, Z. Zhong, X. Xie, Q. Yu, and J. Liu, “CT scans of patients with 2019 novel coronavirus (COVID-19) pneumonia,” Theranostics, vol. 10, no. 10, pp. 4606–4613, 2020, doi: 10.7150/thno.45016.
D. Zhao et al., “A comparative study on the clinical features of coronavirus 2019 (COVID-19) pneumonia with other pneumonias,” Clin. Infect. Dis., vol. 71, no. 15, pp. 756–761, Jul. 2020, doi: 10.1093/cid/ciaa247.
J. Zhao, Y. Zhang, X. He, and P. Xie, “COVID-CT-Dataset: a CT scan dataset about COVID-19,” arXiv Prepr. arXiv2003.13865, 2020. Available: arXiv.
K. Zhang et al., “Clinically applicable AI System for accurate diagnosis, quantitative measurements, and prognosis of COVID-19 pneumonia using computed tomography,” Cell, vol. 181, no. 6, pp. 1423-1433.e11, Jun. 2020, doi: 10.1016/j.cell.2020.04.045.
Y. Song et al., “Deep learning enables accurate diagnosis of novel coronavirus (COVID-19) with CT images,” medRxiv, doi: 10.1101/2020.02.23.20026930.
C. Butt, J. Gill, D. Chun, and B. A. Babu, “Deep learning system to screen coronavirus disease 2019 pneumonia,” Appl. Intell., pp. 1–7, Apr. 2020, doi: 10.1007/s10489-020-01714-3.
L. Li et al., “Using artificial intelligence to detect COVID-19 and community-acquired pneumonia based on pulmonary CT: Evaluation of the diagnostic accuracy,” Radiology, vol. 296, no. 2, pp. E65–E71, Aug. 2020, doi: 10.1148/radiol.2020200905.
H. Gunraj, L. Wang, and A. Wong, “COVIDNet-CT: A tailored deep convolutional neural network design for detection of COVID-19 cases from chest CT images,” arXiv Prepr. arXiv2009.05383, 2020. Available: arXiv
Y.-H. Wu et al., “JCS: An explainable COVID-19 diagnosis system by joint classification and segmentation,” arXiv Prepr. arXiv2004.07054, 2020. Available: arXiv
K. He, G. Gkioxari, P. Dollar, and R. Girshick, “Mask R-CNN,” in 2017 IEEE International Conference on Computer Vision (ICCV), Oct. 2017, pp. 2980–2988, doi: 10.1109/ICCV.2017.322.
A. Ter-Sarkisov, “COVID-CT-Mask-Net: Prediction of COVID-19 from CT scans using regional features,” medRxiv, 2020, doi: 10.1101/2020.10.11.20211052.
T.-Y. Lin et al., “Microsoft COCO: Common objects in context,” Eur. Conf. Comput. Vis., pp. 740–755, 2014, doi: 10.1007/978-3-319-10602-1_48.
M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman, “The pascal visual object classes (VOC) challenge,” Int. J. Comput. Vis., vol. 88, no. 2, pp. 303–338, Jun. 2010, doi: 10.1007/s11263-009-0275-4.
Refbacks
- There are currently no refbacks.
Copyright (c) 2020 Aram Ter-Sarkisov
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
___________________________________________________________
Science in Information Technology Letters
ISSN 2722-4139
Published by Association for Scientific Computing Electrical and Engineering (ASCEE)
W : http://pubs2.ascee.org/index.php/sitech
E : sitech@ascee.org, andri@ascee.org, andri.pranolo.id@ieee.org
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0