Enhancing the stability of CT radiomics across different volume of interest sizes using parametric feature maps: a phantom study.
Radiomics
Reproducibility of results
Texture analysis
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:
15 09 2022
15 09 2022
Historique:
received:
26
05
2022
accepted:
05
07
2022
entrez:
14
9
2022
pubmed:
15
9
2022
medline:
17
9
2022
Statut:
epublish
Résumé
In radiomics studies, differences in the volume of interest (VOI) are often inevitable and may confound the extracted features. We aimed to correct this confounding effect of VOI variability by applying parametric maps with a fixed voxel size. Ten scans of a cup filled with sodium chloride solution were scanned using a multislice computed tomography (CT) unit. Sphere-shaped VOIs with different diameters (4, 8, or 16 mm) were drawn centrally into the phantom. A total of 93 features were extracted conventionally from the original images using PyRadiomics. Using a self-designed and pretested software tool, parametric maps for the same 93 features with a fixed voxel size of 4 mm Fifty-five conventionally extracted features were significantly different between the VOI sizes, and none of the features showed excellent agreement in terms of OCCCs. When read from the parametric maps, only 8 features showed significant differences, and 3 features showed an excellent OCCC (≥ 0.85). The OCCCs for 89 features substantially increased using the parametric maps. This phantom study shows that converting CT images into parametric maps resolves the confounding effect of VOI variability and increases feature reproducibility across VOI sizes.
Sections du résumé
BACKGROUND
In radiomics studies, differences in the volume of interest (VOI) are often inevitable and may confound the extracted features. We aimed to correct this confounding effect of VOI variability by applying parametric maps with a fixed voxel size.
METHODS
Ten scans of a cup filled with sodium chloride solution were scanned using a multislice computed tomography (CT) unit. Sphere-shaped VOIs with different diameters (4, 8, or 16 mm) were drawn centrally into the phantom. A total of 93 features were extracted conventionally from the original images using PyRadiomics. Using a self-designed and pretested software tool, parametric maps for the same 93 features with a fixed voxel size of 4 mm
RESULTS
Fifty-five conventionally extracted features were significantly different between the VOI sizes, and none of the features showed excellent agreement in terms of OCCCs. When read from the parametric maps, only 8 features showed significant differences, and 3 features showed an excellent OCCC (≥ 0.85). The OCCCs for 89 features substantially increased using the parametric maps.
CONCLUSIONS
This phantom study shows that converting CT images into parametric maps resolves the confounding effect of VOI variability and increases feature reproducibility across VOI sizes.
Identifiants
pubmed: 36104519
doi: 10.1186/s41747-022-00297-7
pii: 10.1186/s41747-022-00297-7
pmc: PMC9474978
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
43Informations de copyright
© 2022. The Author(s) under exclusive licence to European Society of Radiology.
Références
Eur Radiol Exp. 2022 Jan 25;6(1):2
pubmed: 35075539
Radiology. 2018 Aug;288(2):407-415
pubmed: 29688159
Radiology. 2016 Feb;278(2):563-77
pubmed: 26579733
Transl Oncol. 2014 Feb 01;7(1):72-87
pubmed: 24772210
Tomography. 2021 Sep 17;7(3):477-487
pubmed: 34564303
Radiother Oncol. 2008 Apr;87(1):142-6
pubmed: 18207269
Med Phys. 2017 Mar;44(3):1050-1062
pubmed: 28112418
Eur Radiol Exp. 2022 May 3;6(1):19
pubmed: 35501512
Eur Radiol. 2021 Mar;31(3):1460-1470
pubmed: 32909055
Cancer Res. 2017 Nov 1;77(21):e104-e107
pubmed: 29092951
Eur Radiol Exp. 2018 Nov 14;2(1):36
pubmed: 30426318
Int J Med Phys Clin Eng Radiat Oncol. 2018 Aug;7(3):330-338
pubmed: 31131158
Tomography. 2020 Jun;6(2):118-128
pubmed: 32548288
Invest Radiol. 2015 Nov;50(11):757-65
pubmed: 26115366
Sci Rep. 2018 Jul 12;8(1):10545
pubmed: 30002441
Front Oncol. 2021 Nov 17;11:773512
pubmed: 34869015
Acta Oncol. 2017 Nov;56(11):1544-1553
pubmed: 28885084
Invest Radiol. 2019 Apr;54(4):221-228
pubmed: 30433891
Eur Radiol Exp. 2020 Nov 10;4(1):62
pubmed: 33169295
Front Oncol. 2021 May 27;11:638185
pubmed: 34123789
Invest Radiol. 2020 Sep;55(9):601-616
pubmed: 32209816
EBioMedicine. 2020 Sep;59:102963
pubmed: 32891051
Sci Rep. 2020 Mar 2;10(1):3852
pubmed: 32123281
Radiology. 2020 May;295(2):328-338
pubmed: 32154773
Eur J Radiol. 2013 Feb;82(2):342-8
pubmed: 23194641
Eur Radiol Exp. 2021 Jan 19;5(1):2
pubmed: 33462642
Eur J Cancer. 2012 Mar;48(4):441-6
pubmed: 22257792
Int J Radiat Oncol Biol Phys. 2018 Nov 15;102(4):1143-1158
pubmed: 30170872
Tomography. 2021 Jun 08;7(2):238-252
pubmed: 34201012
Biometrics. 2002 Dec;58(4):1020-7
pubmed: 12495158
Phys Med. 2020 Mar;71:24-30
pubmed: 32088562
Nat Commun. 2014 Jun 03;5:4006
pubmed: 24892406
Biometrics. 1989 Mar;45(1):255-68
pubmed: 2720055
Eur Radiol. 2021 Jan;31(1):1-4
pubmed: 32767103
Nat Rev Clin Oncol. 2017 Dec;14(12):749-762
pubmed: 28975929
Eur Radiol. 2022 Apr 30;:
pubmed: 35499565