Accurate Tumor Delineation

machine learning non-small cell lung cancer prognosis radiomics segmentation

Journal

Frontiers in oncology
ISSN: 2234-943X
Titre abrégé: Front Oncol
Pays: Switzerland
ID NLM: 101568867

Informations de publication

Date de publication:
2021
Historique:
received: 17 06 2021
accepted: 20 09 2021
entrez: 4 11 2021
pubmed: 5 11 2021
medline: 5 11 2021
Statut: epublish

Résumé

The aim of this work was to investigate the ability of building prognostic models in non-small cell lung cancer (NSCLC) using radiomic features from positron emission tomography and computed tomography with 2-deoxy-2-[fluorine-18]fluoro-d-glucose ( A cohort of 138 patients with stage II-III NSCLC treated with radiochemotherapy recruited retrospectively ( Overall, better performance was achieved using the features from delineated tumor volumes. This was observed consistently across ML algorithms and for the two clinical endpoints. However, the loss of performance was not significant, especially when a consensus of the three ML algorithms was considered (0.89 Our findings suggest that it is feasible to achieve similar levels of prognostic accuracy in radiomics-based modeling by relying on a faster and easier VOI definition, skipping a time-consuming tumor delineation step, thus facilitating automation of the whole radiomics workflow. The associated cost is a loss of performance in the resulting models, although this loss can be greatly mitigated when a consensus of several models is relied upon.

Sections du résumé

BACKGROUND BACKGROUND
The aim of this work was to investigate the ability of building prognostic models in non-small cell lung cancer (NSCLC) using radiomic features from positron emission tomography and computed tomography with 2-deoxy-2-[fluorine-18]fluoro-d-glucose (
METHODS METHODS
A cohort of 138 patients with stage II-III NSCLC treated with radiochemotherapy recruited retrospectively (
RESULTS RESULTS
Overall, better performance was achieved using the features from delineated tumor volumes. This was observed consistently across ML algorithms and for the two clinical endpoints. However, the loss of performance was not significant, especially when a consensus of the three ML algorithms was considered (0.89
CONCLUSION CONCLUSIONS
Our findings suggest that it is feasible to achieve similar levels of prognostic accuracy in radiomics-based modeling by relying on a faster and easier VOI definition, skipping a time-consuming tumor delineation step, thus facilitating automation of the whole radiomics workflow. The associated cost is a loss of performance in the resulting models, although this loss can be greatly mitigated when a consensus of several models is relied upon.

Identifiants

pubmed: 34733779
doi: 10.3389/fonc.2021.726865
pmc: PMC8560021
doi:

Types de publication

Journal Article

Langues

eng

Pagination

726865

Informations de copyright

Copyright © 2021 Sepehri, Tankyevych, Iantsen, Visvikis, Hatt and Cheze Le Rest.

Déclaration de conflit d'intérêts

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Références

IEEE Trans Med Imaging. 2009 Jun;28(6):881-93
pubmed: 19150782
Int J Radiat Oncol Biol Phys. 2010 May 1;77(1):301-8
pubmed: 20116934
Radiology. 2020 May;295(2):328-338
pubmed: 32154773
Eur Radiol. 2019 Sep;29(9):4765-4775
pubmed: 30747300
Eur J Nucl Med Mol Imaging. 2013 Oct;40(11):1662-71
pubmed: 23857457
Quant Imaging Med Surg. 2019 Mar;9(3):453-464
pubmed: 31032192
Sci Rep. 2020 Jun 24;10(1):10248
pubmed: 32581221
CA Cancer J Clin. 2018 Nov;68(6):394-424
pubmed: 30207593
Eur J Nucl Med Mol Imaging. 2016 Jul;43(8):1477-85
pubmed: 26896298
Methods. 2021 Apr;188:73-83
pubmed: 33197567
Med Phys. 2019 Aug;46(8):3582-3591
pubmed: 31131906
Br J Radiol. 2021 Jun 1;94(1122):20201272
pubmed: 33882244
Eur J Nucl Med Mol Imaging. 2015 Apr;42(4):634-43
pubmed: 25573632
Eur Radiol. 2021 Feb;31(2):1049-1058
pubmed: 32809167
Eur Radiol. 2020 Jan;30(1):523-536
pubmed: 31350588
Mol Imaging Biol. 2016 Oct;18(5):788-95
pubmed: 26920355
Eur J Nucl Med Mol Imaging. 2018 Apr;45(4):630-641
pubmed: 29177871
J Nucl Med. 2019 Sep;60(Suppl 2):38S-44S
pubmed: 31481588
Eur J Nucl Med Mol Imaging. 2021 Oct;48(11):3444-3456
pubmed: 33772335
Semin Nucl Med. 2021 Mar;51(2):126-133
pubmed: 33509369
Acta Oncol. 2013 Oct;52(7):1391-7
pubmed: 24047337
EBioMedicine. 2019 Sep;47:20-21
pubmed: 31474549
Diagnostics (Basel). 2021 Apr 09;11(4):
pubmed: 33918681
PLoS One. 2014 Jul 15;9(7):e102107
pubmed: 25025374
Future Oncol. 2010 Mar;6(3):367-80
pubmed: 20222794
Radiother Oncol. 2016 Sep;120(3):512-518
pubmed: 27569847
Phys Med Biol. 2020 Dec 17;65(24):24TR02
pubmed: 32688357
Med Phys. 2019 Feb;46(2):665-678
pubmed: 30506687
Mol Imaging Biol. 2019 Dec;21(6):1157-1164
pubmed: 30850967

Auteurs

Shima Sepehri (S)

LaTIM, INSERM, UMR 1101, Univ Brest, Brest, France.

Olena Tankyevych (O)

LaTIM, INSERM, UMR 1101, Univ Brest, Brest, France.
University Hospital Poitiers, Nuclear Medicine Department, Poitiers, France.

Andrei Iantsen (A)

LaTIM, INSERM, UMR 1101, Univ Brest, Brest, France.

Dimitris Visvikis (D)

LaTIM, INSERM, UMR 1101, Univ Brest, Brest, France.

Mathieu Hatt (M)

LaTIM, INSERM, UMR 1101, Univ Brest, Brest, France.

Catherine Cheze Le Rest (C)

LaTIM, INSERM, UMR 1101, Univ Brest, Brest, France.
University Hospital Poitiers, Nuclear Medicine Department, Poitiers, France.

Classifications MeSH