Evaluation of artificial intelligence-based autosegmentation for a high-performance cone-beam computed tomography imaging system in the pelvic region

Background and purpose: A novel ring-gantry cone-beam computed tomography (CBCT) imaging system shows improved image quality compared to its conventional version, but its effect on autosegmentation is unknown. This study evaluates the impact of this high-performance CBCT on autosegmentation performa...

Full description

Saved in:
Bibliographic Details
Main Authors: Judith H. Sluijter, Agustinus J.A.J. van de Schoot, Abdelmounaim el Yaakoubi, Maartje de Jong, Martine S. van der Knaap - van Dongen, Britt Kunnen, Nienke D. Sijtsema, Joan J. Penninkhof, Kim C. de Vries, Steven F. Petit, Maarten L.P. Dirkx
Format: Article
Language:English
Published: Elsevier 2025-01-01
Series:Physics and Imaging in Radiation Oncology
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S240563162400157X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846115928795774976
author Judith H. Sluijter
Agustinus J.A.J. van de Schoot
Abdelmounaim el Yaakoubi
Maartje de Jong
Martine S. van der Knaap - van Dongen
Britt Kunnen
Nienke D. Sijtsema
Joan J. Penninkhof
Kim C. de Vries
Steven F. Petit
Maarten L.P. Dirkx
author_facet Judith H. Sluijter
Agustinus J.A.J. van de Schoot
Abdelmounaim el Yaakoubi
Maartje de Jong
Martine S. van der Knaap - van Dongen
Britt Kunnen
Nienke D. Sijtsema
Joan J. Penninkhof
Kim C. de Vries
Steven F. Petit
Maarten L.P. Dirkx
author_sort Judith H. Sluijter
collection DOAJ
description Background and purpose: A novel ring-gantry cone-beam computed tomography (CBCT) imaging system shows improved image quality compared to its conventional version, but its effect on autosegmentation is unknown. This study evaluates the impact of this high-performance CBCT on autosegmentation performance, inter-observer variability, contour correction times and delineation confidence, compared to the conventional CBCT. Materials and methods: Twenty prostate cancer patients were enrolled in this prospective clinical study. Per patient, one pair of high-performance CBCT and conventional CBCT scans was included. Three observers manually corrected contours generated by the artificial intelligence (AI) model for prostate, seminal vesicles, bladder, rectum and bowel. Differences between AI-based and manual corrected contours were quantified using Dice Similarity Coefficient (DSC) and 95th percentile of Hausdorff distance (HD95). Autosegmentation performance and interobserver variation were compared using a random effects model; correction times and confidence scores using a paired t-test and Wilcoxon signed-rank test, respectively. Results: Autosegmentation performance showed small, but statistically insignificant differences. Interobserver variability, assessed by the intraclass correlation coefficient, was significantly different across most organs, but these were considered clinically irrelevant (maximum difference = 0.08). Mean contour correction times were similar for both CBCT systems (11:03 versus 11:12 min; p = 0.66). Delineation confidence scores were significantly higher with the high-performance CBCT scans for prostate, seminal vesicles and rectum (4.5 versus 3.5, 4.3 versus 3.5, 4.8 versus 4.3; all p < 0.001). Conclusion: The high-performance CBCT did not (clinically) improve autosegmentation performance, inter-observer variability or contour correction time compared to conventional CBCT. However, it clearly enhanced user confidence in organ delineation for prostate, seminal vesicles and rectum.
format Article
id doaj-art-053b08774c57440494076047e6b4c240
institution Kabale University
issn 2405-6316
language English
publishDate 2025-01-01
publisher Elsevier
record_format Article
series Physics and Imaging in Radiation Oncology
spelling doaj-art-053b08774c57440494076047e6b4c2402024-12-19T10:55:45ZengElsevierPhysics and Imaging in Radiation Oncology2405-63162025-01-0133100687Evaluation of artificial intelligence-based autosegmentation for a high-performance cone-beam computed tomography imaging system in the pelvic regionJudith H. Sluijter0Agustinus J.A.J. van de Schoot1Abdelmounaim el Yaakoubi2Maartje de Jong3Martine S. van der Knaap - van Dongen4Britt Kunnen5Nienke D. Sijtsema6Joan J. Penninkhof7Kim C. de Vries8Steven F. Petit9Maarten L.P. Dirkx10Corresponding author.; Department of Radiotherapy, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsDepartment of Radiotherapy, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsDepartment of Radiotherapy, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsDepartment of Radiotherapy, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsDepartment of Radiotherapy, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsDepartment of Radiotherapy, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsDepartment of Radiotherapy, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsDepartment of Radiotherapy, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsDepartment of Radiotherapy, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsDepartment of Radiotherapy, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsDepartment of Radiotherapy, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsBackground and purpose: A novel ring-gantry cone-beam computed tomography (CBCT) imaging system shows improved image quality compared to its conventional version, but its effect on autosegmentation is unknown. This study evaluates the impact of this high-performance CBCT on autosegmentation performance, inter-observer variability, contour correction times and delineation confidence, compared to the conventional CBCT. Materials and methods: Twenty prostate cancer patients were enrolled in this prospective clinical study. Per patient, one pair of high-performance CBCT and conventional CBCT scans was included. Three observers manually corrected contours generated by the artificial intelligence (AI) model for prostate, seminal vesicles, bladder, rectum and bowel. Differences between AI-based and manual corrected contours were quantified using Dice Similarity Coefficient (DSC) and 95th percentile of Hausdorff distance (HD95). Autosegmentation performance and interobserver variation were compared using a random effects model; correction times and confidence scores using a paired t-test and Wilcoxon signed-rank test, respectively. Results: Autosegmentation performance showed small, but statistically insignificant differences. Interobserver variability, assessed by the intraclass correlation coefficient, was significantly different across most organs, but these were considered clinically irrelevant (maximum difference = 0.08). Mean contour correction times were similar for both CBCT systems (11:03 versus 11:12 min; p = 0.66). Delineation confidence scores were significantly higher with the high-performance CBCT scans for prostate, seminal vesicles and rectum (4.5 versus 3.5, 4.3 versus 3.5, 4.8 versus 4.3; all p < 0.001). Conclusion: The high-performance CBCT did not (clinically) improve autosegmentation performance, inter-observer variability or contour correction time compared to conventional CBCT. However, it clearly enhanced user confidence in organ delineation for prostate, seminal vesicles and rectum.http://www.sciencedirect.com/science/article/pii/S240563162400157XAI-based autosegmentationCBCT-guided online adaptive radiotherapyPelvic regionRandom effects model
spellingShingle Judith H. Sluijter
Agustinus J.A.J. van de Schoot
Abdelmounaim el Yaakoubi
Maartje de Jong
Martine S. van der Knaap - van Dongen
Britt Kunnen
Nienke D. Sijtsema
Joan J. Penninkhof
Kim C. de Vries
Steven F. Petit
Maarten L.P. Dirkx
Evaluation of artificial intelligence-based autosegmentation for a high-performance cone-beam computed tomography imaging system in the pelvic region
Physics and Imaging in Radiation Oncology
AI-based autosegmentation
CBCT-guided online adaptive radiotherapy
Pelvic region
Random effects model
title Evaluation of artificial intelligence-based autosegmentation for a high-performance cone-beam computed tomography imaging system in the pelvic region
title_full Evaluation of artificial intelligence-based autosegmentation for a high-performance cone-beam computed tomography imaging system in the pelvic region
title_fullStr Evaluation of artificial intelligence-based autosegmentation for a high-performance cone-beam computed tomography imaging system in the pelvic region
title_full_unstemmed Evaluation of artificial intelligence-based autosegmentation for a high-performance cone-beam computed tomography imaging system in the pelvic region
title_short Evaluation of artificial intelligence-based autosegmentation for a high-performance cone-beam computed tomography imaging system in the pelvic region
title_sort evaluation of artificial intelligence based autosegmentation for a high performance cone beam computed tomography imaging system in the pelvic region
topic AI-based autosegmentation
CBCT-guided online adaptive radiotherapy
Pelvic region
Random effects model
url http://www.sciencedirect.com/science/article/pii/S240563162400157X
work_keys_str_mv AT judithhsluijter evaluationofartificialintelligencebasedautosegmentationforahighperformanceconebeamcomputedtomographyimagingsysteminthepelvicregion
AT agustinusjajvandeschoot evaluationofartificialintelligencebasedautosegmentationforahighperformanceconebeamcomputedtomographyimagingsysteminthepelvicregion
AT abdelmounaimelyaakoubi evaluationofartificialintelligencebasedautosegmentationforahighperformanceconebeamcomputedtomographyimagingsysteminthepelvicregion
AT maartjedejong evaluationofartificialintelligencebasedautosegmentationforahighperformanceconebeamcomputedtomographyimagingsysteminthepelvicregion
AT martinesvanderknaapvandongen evaluationofartificialintelligencebasedautosegmentationforahighperformanceconebeamcomputedtomographyimagingsysteminthepelvicregion
AT brittkunnen evaluationofartificialintelligencebasedautosegmentationforahighperformanceconebeamcomputedtomographyimagingsysteminthepelvicregion
AT nienkedsijtsema evaluationofartificialintelligencebasedautosegmentationforahighperformanceconebeamcomputedtomographyimagingsysteminthepelvicregion
AT joanjpenninkhof evaluationofartificialintelligencebasedautosegmentationforahighperformanceconebeamcomputedtomographyimagingsysteminthepelvicregion
AT kimcdevries evaluationofartificialintelligencebasedautosegmentationforahighperformanceconebeamcomputedtomographyimagingsysteminthepelvicregion
AT stevenfpetit evaluationofartificialintelligencebasedautosegmentationforahighperformanceconebeamcomputedtomographyimagingsysteminthepelvicregion
AT maartenlpdirkx evaluationofartificialintelligencebasedautosegmentationforahighperformanceconebeamcomputedtomographyimagingsysteminthepelvicregion