COLI-Net: Deep learning-assisted fully automated COVID-19 lung and infection pneumonia lesion detection and segmentation from chest computed tomography images.
COVID‐19
X‐ray CT
deep learning
pneumonia
segmentation
Journal
International journal of imaging systems and technology
ISSN: 0899-9457
Titre abrégé: Int J Imaging Syst Technol
Pays: United States
ID NLM: 9883966
Informations de publication
Date de publication:
Jan 2022
Jan 2022
Historique:
received:
10
06
2021
revised:
18
09
2021
accepted:
17
10
2021
pubmed:
14
12
2021
medline:
14
12
2021
entrez:
13
12
2021
Statut:
ppublish
Résumé
We present a deep learning (DL)-based automated whole lung and COVID-19 pneumonia infectious lesions (COLI-Net) detection and segmentation from chest computed tomography (CT) images. This multicenter/multiscanner study involved 2368 (347'259 2D slices) and 190 (17 341 2D slices) volumetric CT exams along with their corresponding manual segmentation of lungs and lesions, respectively. All images were cropped, resized, and the intensity values clipped and normalized. A residual network with non-square Dice loss function built upon TensorFlow was employed. The accuracy of lung and COVID-19 lesions segmentation was evaluated on an external reverse transcription-polymerase chain reaction positive COVID-19 dataset (7'333 2D slices) collected at five different centers. To evaluate the segmentation performance, we calculated different quantitative metrics, including radiomic features. The mean Dice coefficients were 0.98 ± 0.011 (95% CI, 0.98-0.99) and 0.91 ± 0.038 (95% CI, 0.90-0.91) for lung and lesions segmentation, respectively. The mean relative Hounsfield unit differences were 0.03 ± 0.84% (95% CI, -0.12 to 0.18) and -0.18 ± 3.4% (95% CI, -0.8 to 0.44) for the lung and lesions, respectively. The relative volume difference for lung and lesions were 0.38 ± 1.2% (95% CI, 0.16-0.59) and 0.81 ± 6.6% (95% CI, -0.39 to 2), respectively. Most radiomic features had a mean relative error less than 5% with the highest mean relative error achieved for the lung for the
Identifiants
pubmed: 34898850
doi: 10.1002/ima.22672
pii: IMA22672
pmc: PMC8652855
doi:
Types de publication
Journal Article
Langues
eng
Pagination
12-25Informations de copyright
© 2021 The Authors. International Journal of Imaging Systems and Technology published by Wiley Periodicals LLC.
Déclaration de conflit d'intérêts
The authors declare no conflicts of interest.
Références
Eur J Nucl Med Mol Imaging. 2021 Mar;48(3):670-682
pubmed: 32875430
SN Comput Sci. 2022;3(1):13
pubmed: 34723206
J Digit Imaging. 2016 Aug;29(4):507-19
pubmed: 26961983
Radiology. 2021 Jan;298(1):E18-E28
pubmed: 32729810
IEEE Trans Med Imaging. 2005 Aug;24(8):1025-38
pubmed: 16092334
J Nucl Cardiol. 2021 Dec;28(6):2761-2779
pubmed: 32347527
IEEE Rev Biomed Eng. 2021;14:4-15
pubmed: 32305937
IEEE Trans Med Imaging. 2018 Dec;37(12):2663-2674
pubmed: 29994201
Radiology. 2020 Aug;296(2):E32-E40
pubmed: 32101510
J Med Syst. 2018 Oct 8;42(11):226
pubmed: 30298337
IEEE Trans Med Imaging. 2020 Aug;39(8):2653-2663
pubmed: 32730215
Eur Radiol. 2021 Mar;31(3):1420-1431
pubmed: 32879987
JAMA. 2020 Apr 7;323(13):1239-1242
pubmed: 32091533
Int J Imaging Syst Technol. 2021 Mar;31(1):16-27
pubmed: 33362345
Theranostics. 2020 Jun 5;10(16):7231-7244
pubmed: 32641989
J Digit Imaging. 2013 Dec;26(6):1045-57
pubmed: 23884657
Eur Radiol. 2020 Dec;30(12):6554-6560
pubmed: 32621238
N Engl J Med. 1997 Jan 23;336(4):243-50
pubmed: 8995086
N Engl J Med. 2020 Feb 20;382(8):727-733
pubmed: 31978945
Radiother Oncol. 2021 Jun;159:231-240
pubmed: 33831446
Eur Radiol. 2020 Dec;30(12):6828-6837
pubmed: 32683550
Int J Surg. 2020 Jun;78:185-193
pubmed: 32305533
Radiology. 2020 Aug;296(2):E46-E54
pubmed: 32155105
Eur J Nucl Med Mol Imaging. 2020 Oct;47(11):2533-2548
pubmed: 32415552
Clin Nucl Med. 2021 Nov 1;46(11):872-883
pubmed: 34238799
IEEE Trans Med Imaging. 2020 Aug;39(8):2626-2637
pubmed: 32730213
J Med Internet Res. 2020 Jun 29;22(6):e19569
pubmed: 32568730
Phys Med. 2021 Aug;88:127-137
pubmed: 34242884
Nat Rev Drug Discov. 2020 May;19(5):305-306
pubmed: 32273591
IEEE Trans Med Imaging. 2019 Feb;38(2):540-549
pubmed: 30716024
Cell. 2020 Sep 3;182(5):1360
pubmed: 32888496
BMJ. 2020 Apr 7;369:m1328
pubmed: 32265220
Deep Learn Med Image Anal Multimodal Learn Clin Decis Support (2018). 2018 Sep;11045:3-11
pubmed: 32613207
Med Image Anal. 2019 Apr;53:197-207
pubmed: 30802813
Comput Methods Programs Biomed. 2021 Apr;202:106004
pubmed: 33662804
IEEE Trans Med Imaging. 2020 Aug;39(8):2615-2625
pubmed: 33156775
Nat Med. 2020 Aug;26(8):1224-1228
pubmed: 32427924
J Digit Imaging. 2019 Aug;32(4):582-596
pubmed: 31144149
Int J Imaging Syst Technol. 2022 Jan;32(1):12-25
pubmed: 34898850
J Biomol Struct Dyn. 2021 Sep;39(15):5682-5689
pubmed: 32619398
Eur Respir J. 2020 Aug 6;56(2):
pubmed: 32444412
Radiology. 2020 May;295(2):E3
pubmed: 32142398
IEEE Trans Med Imaging. 2020 Jul;39(7):2531-2540
pubmed: 32070947
Med Phys. 2020 Nov;47(11):5941-5952
pubmed: 32749075
Phys Med. 2021 Mar;83:122-137
pubmed: 33765602
Comput Biol Med. 2021 Sep;136:104755
pubmed: 34388458
JAMA. 2020 May 12;323(18):1824-1836
pubmed: 32282022
Med Phys. 2009 Oct;36(10):4592-9
pubmed: 19928090
Eur Radiol Exp. 2020 Aug 20;4(1):50
pubmed: 32814998
Radiology. 2021 Jan;298(1):E30-E37
pubmed: 32776832
Radiology. 2020 Aug;296(2):E115-E117
pubmed: 32073353
Med Phys. 2018 Oct;45(10):4568-4581
pubmed: 30144101
Radiol Artif Intell. 2020 Jul 29;2(4):e200048
pubmed: 33928255