A multicenter evaluation of a deep learning software (LungQuant) for lung parenchyma characterization in COVID-19 pneumonia.
COVID-19
Deep Learning
Lung
Software validation
Tomography (x-ray computed)
Journal
European radiology experimental
ISSN: 2509-9280
Titre abrégé: Eur Radiol Exp
Pays: England
ID NLM: 101721752
Informations de publication
Date de publication:
10 04 2023
10 04 2023
Historique:
received:
01
11
2022
accepted:
27
02
2023
medline:
11
4
2023
entrez:
9
4
2023
pubmed:
10
4
2023
Statut:
epublish
Résumé
The role of computed tomography (CT) in the diagnosis and characterization of coronavirus disease 2019 (COVID-19) pneumonia has been widely recognized. We evaluated the performance of a software for quantitative analysis of chest CT, the LungQuant system, by comparing its results with independent visual evaluations by a group of 14 clinical experts. The aim of this work is to evaluate the ability of the automated tool to extract quantitative information from lung CT, relevant for the design of a diagnosis support model. LungQuant segments both the lungs and lesions associated with COVID-19 pneumonia (ground-glass opacities and consolidations) and computes derived quantities corresponding to qualitative characteristics used to clinically assess COVID-19 lesions. The comparison was carried out on 120 publicly available CT scans of patients affected by COVID-19 pneumonia. Scans were scored for four qualitative metrics: percentage of lung involvement, type of lesion, and two disease distribution scores. We evaluated the agreement between the LungQuant output and the visual assessments through receiver operating characteristics area under the curve (AUC) analysis and by fitting a nonlinear regression model. Despite the rather large heterogeneity in the qualitative labels assigned by the clinical experts for each metric, we found good agreement on the metrics compared to the LungQuant output. The AUC values obtained for the four qualitative metrics were 0.98, 0.85, 0.90, and 0.81. Visual clinical evaluation could be complemented and supported by computer-aided quantification, whose values match the average evaluation of several independent clinical experts. We conducted a multicenter evaluation of the deep learning-based LungQuant automated software. We translated qualitative assessments into quantifiable metrics to characterize coronavirus disease 2019 (COVID-19) pneumonia lesions. Comparing the software output to the clinical evaluations, results were satisfactory despite heterogeneity of the clinical evaluations. An automatic quantification tool may contribute to improve the clinical workflow of COVID-19 pneumonia.
Sections du résumé
BACKGROUND
The role of computed tomography (CT) in the diagnosis and characterization of coronavirus disease 2019 (COVID-19) pneumonia has been widely recognized. We evaluated the performance of a software for quantitative analysis of chest CT, the LungQuant system, by comparing its results with independent visual evaluations by a group of 14 clinical experts. The aim of this work is to evaluate the ability of the automated tool to extract quantitative information from lung CT, relevant for the design of a diagnosis support model.
METHODS
LungQuant segments both the lungs and lesions associated with COVID-19 pneumonia (ground-glass opacities and consolidations) and computes derived quantities corresponding to qualitative characteristics used to clinically assess COVID-19 lesions. The comparison was carried out on 120 publicly available CT scans of patients affected by COVID-19 pneumonia. Scans were scored for four qualitative metrics: percentage of lung involvement, type of lesion, and two disease distribution scores. We evaluated the agreement between the LungQuant output and the visual assessments through receiver operating characteristics area under the curve (AUC) analysis and by fitting a nonlinear regression model.
RESULTS
Despite the rather large heterogeneity in the qualitative labels assigned by the clinical experts for each metric, we found good agreement on the metrics compared to the LungQuant output. The AUC values obtained for the four qualitative metrics were 0.98, 0.85, 0.90, and 0.81.
CONCLUSIONS
Visual clinical evaluation could be complemented and supported by computer-aided quantification, whose values match the average evaluation of several independent clinical experts.
KEY POINTS
We conducted a multicenter evaluation of the deep learning-based LungQuant automated software. We translated qualitative assessments into quantifiable metrics to characterize coronavirus disease 2019 (COVID-19) pneumonia lesions. Comparing the software output to the clinical evaluations, results were satisfactory despite heterogeneity of the clinical evaluations. An automatic quantification tool may contribute to improve the clinical workflow of COVID-19 pneumonia.
Identifiants
pubmed: 37032383
doi: 10.1186/s41747-023-00334-z
pii: 10.1186/s41747-023-00334-z
pmc: PMC10083148
doi:
Types de publication
Multicenter Study
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
18Informations de copyright
© 2023. The Author(s).
Références
J Digit Imaging. 2020 Feb;33(1):221-230
pubmed: 31152273
J Appl Clin Med Phys. 2015 Jul 08;16(4):202–215
pubmed: 26218991
Acad Radiol. 2020 May;27(5):603-608
pubmed: 32204987
Quant Imaging Med Surg. 2022 Nov;12(11):5156-5170
pubmed: 36330188
Radiol Med. 2021 Feb;126(2):243-249
pubmed: 33044733
J Chiropr Med. 2016 Jun;15(2):155-63
pubmed: 27330520
Pattern Recognit. 2021 Nov;119:108071
pubmed: 34092815
Eur J Radiol. 2020 Oct;131:109209
pubmed: 32810701
Neuroimage Clin. 2019;23:101846
pubmed: 31077984
Radiol Artif Intell. 2020 Mar 25;2(2):e200029
pubmed: 33937821
Eur Radiol. 2021 May;31(5):2819-2824
pubmed: 33051732
Radiology. 2008 Mar;246(3):697-722
pubmed: 18195376
Radiother Oncol. 2020 Dec;153:139-145
pubmed: 32991916
Rheumatology (Oxford). 2022 Apr 11;61(4):1600-1609
pubmed: 34320649
Phys Med. 2021 Jul;87:115-122
pubmed: 34139383
Chest. 2021 Nov;160(5):e427-e494
pubmed: 34270968
Int J Comput Assist Radiol Surg. 2022 Feb;17(2):229-237
pubmed: 34698988
Phys Med. 2021 Feb;82:28-39
pubmed: 33567361
Eur J Radiol Open. 2020;7:100272
pubmed: 33043101
Eur J Radiol. 2021 May;138:109650
pubmed: 33743491
Invest Radiol. 2020 Jul;55(7):412-421
pubmed: 32304402
Radiol Med. 2020 Jul;125(7):636-646
pubmed: 32500509
Int J Med Sci. 2021 Aug 1;18(15):3395-3402
pubmed: 34522166
Radiology. 2020 Aug;296(2):E113-E114
pubmed: 32105562
Ann Transl Med. 2020 May;8(10):622
pubmed: 32566559
J Infect. 2020 Apr;80(4):394-400
pubmed: 32109443
Emerg Radiol. 2020 Dec;27(6):701-710
pubmed: 33119835
Sci Rep. 2016 Mar 22;6:23376
pubmed: 27001047