Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

AI Improves Agreement and Reduces Time for Quantifying Metabolic Tumour Burden in Hodgkin Lymphoma †

Sadik, May ; Barrington, Sally F. ; Ulén, Johannes ; Enqvist, Olof ; Trägårdh, Elin LU orcid ; Saboury, Babak ; Lerberg Nielsen, Anne ; Loft, Annika ; Loaiza Gongora, Jose Luis and Lopez Urdaneta, Jesus , et al. (2025) In Hematology Reports 17(6).
Abstract

Background: The aim was to evaluate whether an artificial intelligence (AI)-based tool for the automated quantification of the total metabolic tumour volume (tMTV) in patients with Hodgkin lymphoma (HL) could support nuclear medicine specialists in lesion segmentation and thereby enhance inter-observer agreement. Methods: Forty-eight consecutive patients who underwent staging with [18F]FDG PET/CT were included. Eight invited specialists from different hospitals were asked to manually segment lesions for tMTV calculations in 12 cases without AI advice, and to use automated AI segmentation in a further 12 cases, with editing as required, i.e., segmenting/adjusting 24 cases each. Each case was segmented by two specialists manually and by... (More)

Background: The aim was to evaluate whether an artificial intelligence (AI)-based tool for the automated quantification of the total metabolic tumour volume (tMTV) in patients with Hodgkin lymphoma (HL) could support nuclear medicine specialists in lesion segmentation and thereby enhance inter-observer agreement. Methods: Forty-eight consecutive patients who underwent staging with [18F]FDG PET/CT were included. Eight invited specialists from different hospitals were asked to manually segment lesions for tMTV calculations in 12 cases without AI advice, and to use automated AI segmentation in a further 12 cases, with editing as required, i.e., segmenting/adjusting 24 cases each. Each case was segmented by two specialists manually and by two different specialists using the AI tool, allowing for the pairwise comparison of inter-observer variability. Results: The median difference between two specialists performing manual tMTV segmentations was 26 cm3 (IQR 10–86 cm3) corresponding to 23% (IQR 7–50%) of the median tMTV in the dataset, while the median difference between two specialists tMTV adjustments using AI segmentations was 12 cm3 (IQR 4–39 cm3) corresponding to 9% (IQR 2–21%) (p = 0.023). The median difference in tMTV between measurements with and without AI was 3.3 cm3, corresponding to 2.3% of the median tMTV. Conclusions: An automated AI-based tool can significantly increase agreement among specialists quantifying tMTV in HL patients staged with [18F]FDG PET/CT, without markedly changing the measurements.

(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
keywords
artificial intelligence, Fluorodeoxyglucose F18, Hodgkin disease, observer variation, total metabolic tumour volume
in
Hematology Reports
volume
17
issue
6
article number
60
publisher
MDPI AG
external identifiers
  • scopus:105025671774
  • pmid:41283236
ISSN
2038-8322
DOI
10.3390/hematolrep17060060
language
English
LU publication?
yes
id
53f9e2ec-c9e0-4315-b3ab-67adb318236e
date added to LUP
2026-02-12 10:54:48
date last changed
2026-02-13 03:00:02
@article{53f9e2ec-c9e0-4315-b3ab-67adb318236e,
  abstract     = {{<p>Background: The aim was to evaluate whether an artificial intelligence (AI)-based tool for the automated quantification of the total metabolic tumour volume (tMTV) in patients with Hodgkin lymphoma (HL) could support nuclear medicine specialists in lesion segmentation and thereby enhance inter-observer agreement. Methods: Forty-eight consecutive patients who underwent staging with [18F]FDG PET/CT were included. Eight invited specialists from different hospitals were asked to manually segment lesions for tMTV calculations in 12 cases without AI advice, and to use automated AI segmentation in a further 12 cases, with editing as required, i.e., segmenting/adjusting 24 cases each. Each case was segmented by two specialists manually and by two different specialists using the AI tool, allowing for the pairwise comparison of inter-observer variability. Results: The median difference between two specialists performing manual tMTV segmentations was 26 cm<sup>3</sup> (IQR 10–86 cm<sup>3</sup>) corresponding to 23% (IQR 7–50%) of the median tMTV in the dataset, while the median difference between two specialists tMTV adjustments using AI segmentations was 12 cm<sup>3</sup> (IQR 4–39 cm<sup>3</sup>) corresponding to 9% (IQR 2–21%) (p = 0.023). The median difference in tMTV between measurements with and without AI was 3.3 cm<sup>3</sup>, corresponding to 2.3% of the median tMTV. Conclusions: An automated AI-based tool can significantly increase agreement among specialists quantifying tMTV in HL patients staged with [18F]FDG PET/CT, without markedly changing the measurements.</p>}},
  author       = {{Sadik, May and Barrington, Sally F. and Ulén, Johannes and Enqvist, Olof and Trägårdh, Elin and Saboury, Babak and Lerberg Nielsen, Anne and Loft, Annika and Loaiza Gongora, Jose Luis and Lopez Urdaneta, Jesus and Kumar, Rajender and van Essen, Martijn and Edenbrandt, Lars}},
  issn         = {{2038-8322}},
  keywords     = {{artificial intelligence; Fluorodeoxyglucose F18; Hodgkin disease; observer variation; total metabolic tumour volume}},
  language     = {{eng}},
  number       = {{6}},
  publisher    = {{MDPI AG}},
  series       = {{Hematology Reports}},
  title        = {{AI Improves Agreement and Reduces Time for Quantifying Metabolic Tumour Burden in Hodgkin Lymphoma †}},
  url          = {{http://dx.doi.org/10.3390/hematolrep17060060}},
  doi          = {{10.3390/hematolrep17060060}},
  volume       = {{17}},
  year         = {{2025}},
}