Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Comparison between artificial intelligence-based and manual organ delineations in pretreatment computed tomography scans of prostate cancer patients : a visual grading study

Polymeri, Eirini ; Johnsson, Åse A. ; Enqvist, Olof ; Ulén, Johannes ; Kindblom, Jon ; Braide, Karin ; Wiltz, Hans Jurgen ; Tanyasiová, Margareta ; Trägårdh, Elin LU orcid and Edenbrandt, Lars , et al. (2026) In Radiation Protection Dosimetry 202(3-4). p.204-213
Abstract

This study aimed to evaluate the clinical acceptability of artificial intelligence (AI)-based organ segmentations on pretreatment CT images of prostate cancer patients using manual organ delineations as a reference. Paired AI-based segmentations and manual delineations of the prostate, urinary bladder, and rectum were evaluated by three observers, according to a 4-grade Likert-scale, based on quality criteria, developed through a Delphi process. Visual grading characteristics (VGC) analysis was performed. When comparing the ratings of AI-based (n = 360) and manual delineations (n = 360), the area under the VGC-curve (AUCVGC) was 0.36 (95% CI 0.27–0.44), 0.35 (95% CI 0.28–0.41), and 0.3 (95% CI 0.22–0.40) for the prostate,... (More)

This study aimed to evaluate the clinical acceptability of artificial intelligence (AI)-based organ segmentations on pretreatment CT images of prostate cancer patients using manual organ delineations as a reference. Paired AI-based segmentations and manual delineations of the prostate, urinary bladder, and rectum were evaluated by three observers, according to a 4-grade Likert-scale, based on quality criteria, developed through a Delphi process. Visual grading characteristics (VGC) analysis was performed. When comparing the ratings of AI-based (n = 360) and manual delineations (n = 360), the area under the VGC-curve (AUCVGC) was 0.36 (95% CI 0.27–0.44), 0.35 (95% CI 0.28–0.41), and 0.3 (95% CI 0.22–0.40) for the prostate, urinary bladder, and rectum, respectively, indicating inferior ratings for the algorithm. Few AI segmentations (8%) were considered clinically unacceptable, while in 67% no or minor changes were needed. Despite superior ratings for manual delineations, most AI-segmentations needed no or minor changes, indicating clinical acceptability.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; ; ; and , et al. (More)
; ; ; ; ; ; ; ; ; ; and (Less)
organization
publishing date
type
Contribution to journal
publication status
published
subject
in
Radiation Protection Dosimetry
volume
202
issue
3-4
pages
10 pages
publisher
Oxford University Press
external identifiers
  • scopus:105032771468
  • pmid:41821468
ISSN
0144-8420
DOI
10.1093/rpd/ncaf184
language
English
LU publication?
yes
id
53908703-4041-4991-8d4f-83cb5ce99470
date added to LUP
2026-04-22 14:22:27
date last changed
2026-05-20 16:09:20
@article{53908703-4041-4991-8d4f-83cb5ce99470,
  abstract     = {{<p>This study aimed to evaluate the clinical acceptability of artificial intelligence (AI)-based organ segmentations on pretreatment CT images of prostate cancer patients using manual organ delineations as a reference. Paired AI-based segmentations and manual delineations of the prostate, urinary bladder, and rectum were evaluated by three observers, according to a 4-grade Likert-scale, based on quality criteria, developed through a Delphi process. Visual grading characteristics (VGC) analysis was performed. When comparing the ratings of AI-based (n = 360) and manual delineations (n = 360), the area under the VGC-curve (AUC<sub>VGC</sub>) was 0.36 (95% CI 0.27–0.44), 0.35 (95% CI 0.28–0.41), and 0.3 (95% CI 0.22–0.40) for the prostate, urinary bladder, and rectum, respectively, indicating inferior ratings for the algorithm. Few AI segmentations (8%) were considered clinically unacceptable, while in 67% no or minor changes were needed. Despite superior ratings for manual delineations, most AI-segmentations needed no or minor changes, indicating clinical acceptability.</p>}},
  author       = {{Polymeri, Eirini and Johnsson, Åse A. and Enqvist, Olof and Ulén, Johannes and Kindblom, Jon and Braide, Karin and Wiltz, Hans Jurgen and Tanyasiová, Margareta and Trägårdh, Elin and Edenbrandt, Lars and Kjölhede, Henrik and Svalkvist, Angelica}},
  issn         = {{0144-8420}},
  language     = {{eng}},
  month        = {{03}},
  number       = {{3-4}},
  pages        = {{204--213}},
  publisher    = {{Oxford University Press}},
  series       = {{Radiation Protection Dosimetry}},
  title        = {{Comparison between artificial intelligence-based and manual organ delineations in pretreatment computed tomography scans of prostate cancer patients : a visual grading study}},
  url          = {{http://dx.doi.org/10.1093/rpd/ncaf184}},
  doi          = {{10.1093/rpd/ncaf184}},
  volume       = {{202}},
  year         = {{2026}},
}