Glioblastoma and primary central nervous system lymphoma: differentiation using MRI derived first-order texture analysis - a machine learning study.
MRI
glioblastomas
machine learning
primary CNS lymphoma
texture/radiomics
Journal
The neuroradiology journal
ISSN: 2385-1996
Titre abrégé: Neuroradiol J
Pays: United States
ID NLM: 101295103
Informations de publication
Date de publication:
Aug 2021
Aug 2021
Historique:
pubmed:
5
3
2021
medline:
15
12
2021
entrez:
4
3
2021
Statut:
ppublish
Résumé
To evaluate the diagnostic performance of multiple machine learning classifier models derived from first-order histogram texture parameters extracted from T1-weighted contrast-enhanced images in differentiating glioblastoma and primary central nervous system lymphoma. Retrospective study with 97 glioblastoma and 46 primary central nervous system lymphoma patients. Thirty-six different combinations of classifier models and feature selection techniques were evaluated. Five-fold nested cross-validation was performed. Model performance was assessed for whole tumour and largest single slice using receiver operating characteristic curve. The cross-validated model performance was relatively similar for the top performing models for both whole tumour and largest single slice (area under the curve 0.909-0.924). However, there was a considerable difference between the worst performing model (logistic regression with full feature set, area under the curve 0.737) and the highest performing model for whole tumour (least absolute shrinkage and selection operator model with correlation filter, area under the curve 0.924). For single slice, the multilayer perceptron model with correlation filter had the highest performance (area under the curve 0.914). No significant difference was seen between the diagnostic performance of the top performing model for both whole tumour and largest single slice. T1 contrast-enhanced derived first-order texture analysis can differentiate between glioblastoma and primary central nervous system lymphoma with good diagnostic performance. The machine learning performance can vary significantly depending on the model and feature selection methods. Largest single slice and whole tumour analysis show comparable diagnostic performance.
Identifiants
pubmed: 33657924
doi: 10.1177/1971400921998979
pmc: PMC8447821
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
320-328Références
World Neurosurg. 2019 Mar;123:e303-e309
pubmed: 30502475
Acta Radiol. 2019 Mar;60(3):356-366
pubmed: 29860889
Magn Reson Med Sci. 2018 Jan 10;17(1):50-57
pubmed: 28638001
Eur Radiol. 2018 Sep;28(9):3832-3839
pubmed: 29626238
Neuro Oncol. 2018 Aug 2;20(9):1251-1261
pubmed: 29438500
Abdom Radiol (NY). 2019 Oct;44(10):3336-3349
pubmed: 31300850
Eur J Radiol. 2019 Apr;113:116-123
pubmed: 30927935
Neuroradiology. 2018 Dec;60(12):1297-1305
pubmed: 30232517
PLoS One. 2014 Nov 13;9(11):e112948
pubmed: 25393543
Sci Rep. 2019 Apr 5;9(1):5746
pubmed: 30952930
Cancer Imaging. 2013 Sep 23;13(3):400-6
pubmed: 24061266
Neurosurg Focus. 2018 Nov 1;45(5):E5
pubmed: 30453459
Chin Med Sci J. 2019 Mar 30;34(1):10-17
pubmed: 30961775
Ann Transl Med. 2019 Jun;7(11):232
pubmed: 31317002
Clin Neurol Neurosurg. 2018 Oct;173:84-90
pubmed: 30092408
Magn Reson Imaging. 2019 Nov;63:244-249
pubmed: 31425811
Eur J Radiol. 2018 Nov;108:147-154
pubmed: 30396648
Radiology. 2013 Jan;266(1):177-84
pubmed: 23151829
Br J Cancer. 2011 Oct 25;105(9):1414-8
pubmed: 21915121
AJNR Am J Neuroradiol. 2019 Jun;40(6):928-934
pubmed: 31122918
AJNR Am J Neuroradiol. 2017 Mar;38(3):485-491
pubmed: 27932505
Magn Reson Imaging Clin N Am. 2016 Nov;24(4):719-729
pubmed: 27742112
World Neurosurg. 2018 Nov;119:e890-e897
pubmed: 30099179
Neuroradiology. 2019 Jul;61(7):757-765
pubmed: 30949746
Eur J Radiol. 2019 Sep;118:251-256
pubmed: 31439251
Front Comput Neurosci. 2019 Aug 27;13:58
pubmed: 31507398
CNS Neurol Disord Drug Targets. 2017;16(2):160-168
pubmed: 27758687
Clin Neurol Neurosurg. 2020 Nov;198:106205
pubmed: 32932028
Med Phys. 2020 Jun;47(5):e185-e202
pubmed: 32418336
Eur Radiol. 2019 Mar;29(3):1348-1354
pubmed: 30167811
AJNR Am J Neuroradiol. 2017 Jun;38(6):1145-1150
pubmed: 28450433