Dual-Energy CT Texture Analysis With Machine Learning for the Evaluation and Characterization of Cervical Lymphadenopathy.
AUC, Area under the receiver operating curve
Artificial intelligence
DECT, Dual-energy CT
Dual energy CT
HNSCC, Head and neck squamous cell carcinoma
Lymph nodes
Machine learning
NPV, Negative predictive value
PPV, Positive predictive value
Radiomics
Texture analysis
VMI, Virtual monochromatic image
Journal
Computational and structural biotechnology journal
ISSN: 2001-0370
Titre abrégé: Comput Struct Biotechnol J
Pays: Netherlands
ID NLM: 101585369
Informations de publication
Date de publication:
2019
2019
Historique:
received:
29
11
2018
revised:
09
07
2019
accepted:
10
07
2019
entrez:
14
8
2019
pubmed:
14
8
2019
medline:
14
8
2019
Statut:
epublish
Résumé
To determine whether machine learning assisted-texture analysis of multi-energy virtual monochromatic image (VMI) datasets from dual-energy CT (DECT) can be used to differentiate metastatic head and neck squamous cell carcinoma (HNSCC) lymph nodes from lymphoma, inflammatory, or normal lymph nodes. A retrospective evaluation of 412 cervical nodes from 5 different patient groups (50 patients in total) having undergone DECT of the neck between 2013 and 2015 was performed: (1) HNSCC with pathology proven metastatic adenopathy, (2) HNSCC with pathology proven benign nodes (controls for (1)), (3) lymphoma, (4) inflammatory, and (5) normal nodes (controls for (3) and (4)). Texture analysis was performed with TexRAD® software using two independent sets of contours to assess the impact of inter-rater variation. Two machine learning algorithms (Random Forests (RF) and Gradient Boosting Machine (GBM)) were used with independent training and testing sets and determination of accuracy, sensitivity, specificity, PPV, NPV, and AUC. In the independent testing (prediction) sets, the accuracy for distinguishing different groups of pathologic nodes or normal nodes ranged between 80 and 95%. The models generated using texture data extracted from the independent contour sets had substantial to almost perfect agreement. The accuracy, sensitivity, specificity, PPV, and NPV for correctly classifying a lymph node as malignant (i.e. metastatic HNSCC or lymphoma) versus benign were 92%, 91%, 93%, 95%, 87%, respectively. Machine learning assisted-DECT texture analysis can help distinguish different nodal pathology and normal nodes with a high accuracy.
Identifiants
pubmed: 31406557
doi: 10.1016/j.csbj.2019.07.004
pii: S2001-0370(18)30309-X
pmc: PMC6682309
doi:
Types de publication
Journal Article
Langues
eng
Pagination
1009-1015Déclaration de conflit d'intérêts
R.F. has acted as consultant and speaker for GE Healthcare and is a founding partner and stockholder of 4Intel Inc. J.L.W. has received speaker's fees from Siemens Healthcare and GE Healthcare. The other authors have no conflicts of interest to declare.
Références
Radiology. 2004 Mar;230(3):720-6
pubmed: 14990838
Radiology. 2011 Apr;259(1):257-62
pubmed: 21330561
Radiology. 1990 Nov;177(2):379-84
pubmed: 2217772
Radiology. 2013 Dec;269(3):801-9
pubmed: 23912620
Radiol Clin North Am. 2013 Sep;51(5):881-93
pubmed: 24010911
J Comput Assist Tomogr. 2013 Sep-Oct;37(5):666-72
pubmed: 24045238
Eur Radiol. 2014 Mar;24(3):574-80
pubmed: 24081649
Conf Proc IEEE Eng Med Biol Soc. 2013;2013:3973-6
pubmed: 24110602
Radiology. 2014 Jul;272(1):100-12
pubmed: 24654970
Nat Commun. 2014 Jun 03;5:4006
pubmed: 24892406
J Digit Imaging. 2014 Dec;27(6):824-32
pubmed: 24994547
Eur Radiol. 2015 Feb;25(2):480-7
pubmed: 25216770
AJNR Am J Neuroradiol. 2015 Jan;36(1):166-70
pubmed: 25258367
Expert Rev Anticancer Ther. 2015 Feb;15(2):207-24
pubmed: 25385488
Eur J Radiol. 2015 Feb;84(2):312-7
pubmed: 25487819
Invest Radiol. 2015 Apr;50(4):239-45
pubmed: 25501017
Radiology. 2015 Apr;275(1):167-76
pubmed: 25521777
Head Neck. 2016 Apr;38(4):628-34
pubmed: 25524256
Acta Otolaryngol. 2015 Jul;135(7):722-8
pubmed: 25719763
AJNR Am J Neuroradiol. 2015 Jun;36(6):1194-200
pubmed: 25742986
AJNR Am J Neuroradiol. 2015 Jul;36(7):1343-8
pubmed: 25836725
AJNR Am J Neuroradiol. 2015 Aug;36(8):1518-24
pubmed: 26021623
Phys Med Biol. 2015 Jul 21;60(14):5471-96
pubmed: 26119045
Acta Oncol. 2015;54(9):1423-9
pubmed: 26264429
Acta Radiol. 2016 Jun;57(6):669-76
pubmed: 26271125
Sci Rep. 2015 Aug 17;5:13087
pubmed: 26278466
Radiology. 2015 Sep;276(3):637-53
pubmed: 26302388
Expert Rev Anticancer Ther. 2015;15(12):1489-501
pubmed: 26535613
Cancers (Basel). 2015 Nov 06;7(4):2201-16
pubmed: 26561835
J Magn Reson Imaging. 2016 Aug;44(2):445-55
pubmed: 26778191
Sci Rep. 2016 Aug 08;6:31020
pubmed: 27498560
J Comput Assist Tomogr. 2017 Jul/Aug;41(4):565-571
pubmed: 28471869
Radiology. 2017 Sep;284(3):748-757
pubmed: 28493790
Neuroimaging Clin N Am. 2017 Aug;27(3):371-384
pubmed: 28711199
Neuroimaging Clin N Am. 2017 Aug;27(3):385-400
pubmed: 28711200
Neuroimaging Clin N Am. 2017 Aug;27(3):445-459
pubmed: 28711204
Neuroimaging Clin N Am. 2017 Aug;27(3):461-468
pubmed: 28711205
Neuroimaging Clin N Am. 2017 Aug;27(3):533-546
pubmed: 28711211
J Thorac Dis. 2017 Nov;9(11):4758-4766
pubmed: 29268547
Eur Radiol. 2018 Jun;28(6):2604-2611
pubmed: 29294157
AJNR Am J Neuroradiol. 2019 Mar;40(3):543-550
pubmed: 30792253
Eur Radiol. 2019 Apr 12;:null
pubmed: 30980127