Multiphase CT-based prediction of Child-Pugh classification: a machine learning approach.
Artificial intelligence
Liver cirrhosis
Machine learning
Neural networks (computer)
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:
06 04 2020
06 04 2020
Historique:
received:
04
09
2019
accepted:
18
02
2020
entrez:
7
4
2020
pubmed:
7
4
2020
medline:
5
5
2021
Statut:
epublish
Résumé
To evaluate whether machine learning algorithms allow the prediction of Child-Pugh classification on clinical multiphase computed tomography (CT). A total of 259 patients who underwent diagnostic abdominal CT (unenhanced, contrast-enhanced arterial, and venous phases) were included in this retrospective study. Child-Pugh scores were determined based on laboratory and clinical parameters. Linear regression (LR), Random Forest (RF), and convolutional neural network (CNN) algorithms were used to predict the Child-Pugh class. Their performances were compared to the prediction of experienced radiologists (ERs). Spearman correlation coefficients and accuracy were assessed for all predictive models. Additionally, a binary classification in low disease severity (Child-Pugh class A) and advanced disease severity (Child-Pugh class ≥ B) was performed. Eleven imaging features exhibited a significant correlation when adjusted for multiple comparisons with Child-Pugh class. Significant correlations between predicted and measured Child-Pugh classes were observed (ρ The performance of a CNN in assessing Child-Pugh class based on multiphase abdominal CT images is comparable to that of ERs.
Sections du résumé
BACKGROUND
To evaluate whether machine learning algorithms allow the prediction of Child-Pugh classification on clinical multiphase computed tomography (CT).
METHODS
A total of 259 patients who underwent diagnostic abdominal CT (unenhanced, contrast-enhanced arterial, and venous phases) were included in this retrospective study. Child-Pugh scores were determined based on laboratory and clinical parameters. Linear regression (LR), Random Forest (RF), and convolutional neural network (CNN) algorithms were used to predict the Child-Pugh class. Their performances were compared to the prediction of experienced radiologists (ERs). Spearman correlation coefficients and accuracy were assessed for all predictive models. Additionally, a binary classification in low disease severity (Child-Pugh class A) and advanced disease severity (Child-Pugh class ≥ B) was performed.
RESULTS
Eleven imaging features exhibited a significant correlation when adjusted for multiple comparisons with Child-Pugh class. Significant correlations between predicted and measured Child-Pugh classes were observed (ρ
CONCLUSIONS
The performance of a CNN in assessing Child-Pugh class based on multiphase abdominal CT images is comparable to that of ERs.
Identifiants
pubmed: 32249336
doi: 10.1186/s41747-020-00148-3
pii: 10.1186/s41747-020-00148-3
pmc: PMC7131973
doi:
Substances chimiques
Contrast Media
0
Iohexol
4419T9MX03
iopromide
712BAC33MZ
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
20Références
World J Hepatol. 2015 Aug 18;7(17):2069-79
pubmed: 26301049
Genes Dev. 2014 Apr 1;28(7):708-22
pubmed: 24636985
Radiology. 2018 Aug;288(2):407-415
pubmed: 29688159
Ann Hepatol. 2011 May;10 Suppl 1:S6-14
pubmed: 21566249
J Comput Assist Tomogr. 2013 Jul-Aug;37(4):577-82
pubmed: 23863535
Hepatology. 2011 Jun;53(6):2107-10
pubmed: 21547935
QJM. 2010 Jan;103(1):9-16
pubmed: 19903725
J Hepatol. 2005;42 Suppl(1):S100-7
pubmed: 15777564
Cancer Treat Res. 1994;69:293-9
pubmed: 8031659
Magn Reson Imaging. 2012 Nov;30(9):1234-48
pubmed: 22898692
Eur Radiol. 2019 Jan;29(1):383-391
pubmed: 29948090
J Magn Reson Imaging. 2009 Nov;30(5):1042-6
pubmed: 19856436
Invest Radiol. 2015 Apr;50(4):290-6
pubmed: 25493416
AJR Am J Roentgenol. 2007 Jul;189(1):117-22
pubmed: 17579160
Radiology. 2004 Dec;233(3):649-57
pubmed: 15498896
Hepatology. 2008 Mar;47(3):810-6
pubmed: 18098299
Hepatology. 2001 Feb;33(2):473-5
pubmed: 11172352
J Hepatol. 2009 Sep;51(3):433-45
pubmed: 19604596
Radiology. 2018 Apr;287(1):146-155
pubmed: 29239710
Cancer Res. 2017 Nov 1;77(21):e104-e107
pubmed: 29092951
Hepatology. 1993 Jul;18(1):79-85
pubmed: 8392029
Br J Surg. 1973 Aug;60(8):646-9
pubmed: 4541913
Gastroenterology. 2014 Dec;147(6):1338-49.e5; quiz e15
pubmed: 25160979
Biometrics. 1977 Jun;33(2):363-74
pubmed: 884196
Med Image Anal. 2017 Dec;42:60-88
pubmed: 28778026
Intervirology. 2008;51 Suppl 1:17-26
pubmed: 18544944
Ann Pharmacother. 2016 Feb;50(2):113-7
pubmed: 26586854
J Chem Inf Model. 2015 Oct 26;55(10):2085-93
pubmed: 26437739
AJR Am J Roentgenol. 2000 Mar;174(3):705-12
pubmed: 10701613
Hepatology. 2002 Mar;35(3):716-21
pubmed: 11870389
Gastroenterology. 2003 Jan;124(1):91-6
pubmed: 12512033
J Magn Reson Imaging. 2015 Feb;41(2):322-8
pubmed: 24347292