Self-configuring nnU-Net for automatic delineation of the organs at risk and target in high-dose rate cervical brachytherapy, a low/middle-income country's experience.
cervical cancer
deep learning
high-dose rate brachytherapy
segmentation
Journal
Journal of applied clinical medical physics
ISSN: 1526-9914
Titre abrégé: J Appl Clin Med Phys
Pays: United States
ID NLM: 101089176
Informations de publication
Date de publication:
Aug 2023
Aug 2023
Historique:
revised:
01
03
2023
received:
24
11
2022
accepted:
13
03
2023
medline:
7
8
2023
pubmed:
13
4
2023
entrez:
12
4
2023
Statut:
ppublish
Résumé
The high-dose rate (HDR) brachytherapy treatment planning workflow for cervical cancer is a labor-intensive, time-consuming, and expertise-driven process. These issues are amplified in low/middle-income countries with large deficits in experienced healthcare professionals. Automation has the ability to substantially reduce bottlenecks in the planning process but often require a high level of expertise to develop. To implement the out of the box self-configuring nnU-Net package for the auto-segmentation of the organs at risk (OARs) and high-risk CTV (HR CTV) for Ring-Tandem (R-T) HDR cervical brachytherapy treatment planning. The computed tomography (CT) scans of 100 previously treated patients were used to train and test three different nnU-Net configurations (2D, 3DFR, and 3DCasc). The performance of the models was evaluated by calculating the Sørensen-dice similarity coefficient, Hausdorff distance (HD), 95 The mean DSC, HD, HD95, MSD and precision scores for our best performing model (3DFR) were 0.92/7.5 mm/3.0 mm/ 0.8 mm/0.91 for the bladder, 0.84/13.8 mm/5.3 mm/1.4 mm/0.84 for the rectum, and 0.81/8.5 mm/6.0 mm/2.2 mm/0.80 for the HR CTV. Mean dose differences (D Our best performing model (3DFR) provided fast accurate auto generated OARs and HR CTV contours with a large clinical acceptance rate.
Sections du résumé
BACKGROUND
BACKGROUND
The high-dose rate (HDR) brachytherapy treatment planning workflow for cervical cancer is a labor-intensive, time-consuming, and expertise-driven process. These issues are amplified in low/middle-income countries with large deficits in experienced healthcare professionals. Automation has the ability to substantially reduce bottlenecks in the planning process but often require a high level of expertise to develop.
PURPOSE
OBJECTIVE
To implement the out of the box self-configuring nnU-Net package for the auto-segmentation of the organs at risk (OARs) and high-risk CTV (HR CTV) for Ring-Tandem (R-T) HDR cervical brachytherapy treatment planning.
METHODS
METHODS
The computed tomography (CT) scans of 100 previously treated patients were used to train and test three different nnU-Net configurations (2D, 3DFR, and 3DCasc). The performance of the models was evaluated by calculating the Sørensen-dice similarity coefficient, Hausdorff distance (HD), 95
RESULTS
RESULTS
The mean DSC, HD, HD95, MSD and precision scores for our best performing model (3DFR) were 0.92/7.5 mm/3.0 mm/ 0.8 mm/0.91 for the bladder, 0.84/13.8 mm/5.3 mm/1.4 mm/0.84 for the rectum, and 0.81/8.5 mm/6.0 mm/2.2 mm/0.80 for the HR CTV. Mean dose differences (D
CONCLUSION
CONCLUSIONS
Our best performing model (3DFR) provided fast accurate auto generated OARs and HR CTV contours with a large clinical acceptance rate.
Identifiants
pubmed: 37042449
doi: 10.1002/acm2.13988
pmc: PMC10402684
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
e13988Informations de copyright
© 2023 The Authors. Journal of Applied Clinical Medical Physics published by Wiley Periodicals, LLC on behalf of The American Association of Physicists in Medicine.
Références
Radiother Oncol. 2020 Mar;144:152-158
pubmed: 31812930
Brachytherapy. 2014 Mar-Apr;13(2):210-8
pubmed: 24090973
Acta Oncol. 2019 Feb;58(2):257-264
pubmed: 30398090
J Appl Clin Med Phys. 2023 Aug;24(8):e13988
pubmed: 37042449
Nat Methods. 2021 Feb;18(2):203-211
pubmed: 33288961
Phys Med Biol. 2016 Sep 7;61(17):6531-52
pubmed: 27524504
Front Oncol. 2017 Dec 20;7:315
pubmed: 29376025
Radiother Oncol. 2013 Apr;107(1):20-5
pubmed: 23602372
Phys Med. 2018 Jun;50:13-19
pubmed: 29891089
Radiother Oncol. 2006 Jan;78(1):67-77
pubmed: 16403584
Radiother Oncol. 2011 Mar;98(3):373-7
pubmed: 21269714
Int J Radiat Oncol Biol Phys. 1991 Jan;20(1):87-93
pubmed: 1993634
Radiother Oncol. 2013 Apr;107(1):1-5
pubmed: 23541642
Med Phys. 2014 May;41(5):050902
pubmed: 24784366
Med Phys. 2022 Mar;49(3):1571-1584
pubmed: 35094405
Radiother Oncol. 2021 Jun;159:231-240
pubmed: 33831446
Med Image Anal. 2019 May;54:168-178
pubmed: 30928830
Radiother Oncol. 2010 Aug;96(2):153-60
pubmed: 20663578
Cancer. 1983 Mar 1;51(5):959-67
pubmed: 6821861
Brachytherapy. 2012 Jan-Feb;11(1):33-46
pubmed: 22265436
Gynecol Oncol. 2015 Nov;139(2):288-94
pubmed: 26364808
Int J Radiat Oncol Biol Phys. 2021 Mar 15;109(4):1096-1110
pubmed: 33181248
Radiother Oncol. 2012 Apr;103(1):113-22
pubmed: 22296748
Radiother Oncol. 2000 Jul;56(1):37-42
pubmed: 10869753
Med Phys. 2010 Mar;37(3):970-9
pubmed: 20384233
Int J Radiat Oncol Biol Phys. 1991 Apr;20(4):667-76
pubmed: 2004942
Int J Radiat Oncol Biol Phys. 2014 Mar 1;88(3):537-9
pubmed: 24411631
J Med Radiat Sci. 2014 Sep;61(3):151-8
pubmed: 26229651
Lancet Oncol. 2022 Jun;23(6):e251-e312
pubmed: 35550267
Med Image Anal. 2017 Dec;42:60-88
pubmed: 28778026
J Appl Clin Med Phys. 2020 Oct;21(10):158-169
pubmed: 32991783
Int J Radiat Oncol Biol Phys. 2011 Nov 15;81(4):950-7
pubmed: 20932664
Phys Med Biol. 2018 Dec 14;63(24):245015
pubmed: 30523973
Brachytherapy. 2016 Mar-Apr;15(2):191-9
pubmed: 26810408
J Contemp Brachytherapy. 2021 Jun;13(3):325-330
pubmed: 34122573
Radiother Oncol. 2021 Jul;160:273-284
pubmed: 34019918
Radiother Oncol. 2012 Jan;102(1):68-73
pubmed: 21962822
Infect Agent Cancer. 2019 May 21;14:11
pubmed: 31139248
Med Phys. 2022 Mar;49(3):1712-1722
pubmed: 35080018
Int J Radiat Oncol Biol Phys. 2013 Sep 1;87(1):111-9
pubmed: 23849695
Radiother Oncol. 2001 Sep;60(3):273-80
pubmed: 11514007
Cancer. 1990 Dec 15;66(12):2451-6
pubmed: 2249184
Med Phys. 2020 Nov;47(11):5648-5658
pubmed: 32964477
PLoS One. 2017 Oct 6;12(10):e0185844
pubmed: 28985229
Semin Radiat Oncol. 2017 Apr;27(2):184-188
pubmed: 28325246
Med Phys. 2017 Dec;44(12):6377-6389
pubmed: 28963779
J Contemp Brachytherapy. 2019 Oct;11(5):469-478
pubmed: 31749857