Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Validation of an AI Method for Automated Lymphoma Metabolic Tumor Volume Segmentation Using a Public Benchmark PET/CT Dataset

Sadik, May ; Larsson, Måns ; Enqvist, Olof ; Edenbrandt, Lars and Trägårdh, Elin LU orcid (2026) In Journal of nuclear medicine : official publication, Society of Nuclear Medicine 67(6). p.1001-1005
Abstract

The aim of this study was to evaluate the performance of an artificial intelligence (AI)-based method for automated segmentation of total metabolic tumor volume (TMTV) in 18F-FDG PET/CT scans of patients with lymphoma, using an independent, publicly available benchmark dataset curated and segmented by expert readers in a previously published study. Methods: The AI model, based on a 3-dimensional U-Net architecture implemented in MONAI (the medical open-source network for AI framework), was trained on 1,500 18F-FDG PET/CT scans of patients with lymphoma. It was tested on a benchmark dataset comprising 60 baseline scans (20 each of follicular lymphoma, Hodgkin lymphoma, and diffuse large B-cell lymphoma), each segmented by 3 or 4 nuclear... (More)

The aim of this study was to evaluate the performance of an artificial intelligence (AI)-based method for automated segmentation of total metabolic tumor volume (TMTV) in 18F-FDG PET/CT scans of patients with lymphoma, using an independent, publicly available benchmark dataset curated and segmented by expert readers in a previously published study. Methods: The AI model, based on a 3-dimensional U-Net architecture implemented in MONAI (the medical open-source network for AI framework), was trained on 1,500 18F-FDG PET/CT scans of patients with lymphoma. It was tested on a benchmark dataset comprising 60 baseline scans (20 each of follicular lymphoma, Hodgkin lymphoma, and diffuse large B-cell lymphoma), each segmented by 3 or 4 nuclear medicine physicians using an SUV threshold of 4. Agreement between AI-derived and benchmark TMTVs was assessed using Bland-Altman analysis, with acceptable deviation defined as within 10% or 10 cm3, consistent with interreader variability reported in the benchmark study. Results: In 50 (83%) of the 60 benchmark cases, AI-derived TMTVs were within 10% or 10 cm3 of the benchmark reference. In 4 of the remaining 10 cases, AI-derived results were within the same margin of at least 1 of the expert readers, indicating partial concordance. Conclusion: The AI-based method achieved high concordance with expert-derived TMTVs in a standardized benchmark setting. The findings demonstrate that the AI model performs comparably to human experts in most cases, even in an externally curated dataset deliberately enriched with challenging cases by its original authors. The AI model's ability to produce accurate, reproducible segmentations without user interaction could significantly reduce manual workload and interreader variability in lymphoma imaging. However, human supervision is required to minimize errors.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
artificial intelligence, benchmark dataset, deep learning, lymphoma, total metabolic tumor volume
in
Journal of nuclear medicine : official publication, Society of Nuclear Medicine
volume
67
issue
6
pages
5 pages
publisher
Society of Nuclear Medicine Inc.
external identifiers
  • pmid:41786479
  • scopus:105040850349
ISSN
0161-5505
DOI
10.2967/jnumed.125.271605
language
English
LU publication?
yes
id
de20ba57-3729-4fc7-8475-f9f523d978d7
date added to LUP
2026-07-02 13:55:38
date last changed
2026-09-24 20:41:21
@article{de20ba57-3729-4fc7-8475-f9f523d978d7,
  abstract     = {{<p>The aim of this study was to evaluate the performance of an artificial intelligence (AI)-based method for automated segmentation of total metabolic tumor volume (TMTV) in 18F-FDG PET/CT scans of patients with lymphoma, using an independent, publicly available benchmark dataset curated and segmented by expert readers in a previously published study. Methods: The AI model, based on a 3-dimensional U-Net architecture implemented in MONAI (the medical open-source network for AI framework), was trained on 1,500 18F-FDG PET/CT scans of patients with lymphoma. It was tested on a benchmark dataset comprising 60 baseline scans (20 each of follicular lymphoma, Hodgkin lymphoma, and diffuse large B-cell lymphoma), each segmented by 3 or 4 nuclear medicine physicians using an SUV threshold of 4. Agreement between AI-derived and benchmark TMTVs was assessed using Bland-Altman analysis, with acceptable deviation defined as within 10% or 10 cm3, consistent with interreader variability reported in the benchmark study. Results: In 50 (83%) of the 60 benchmark cases, AI-derived TMTVs were within 10% or 10 cm3 of the benchmark reference. In 4 of the remaining 10 cases, AI-derived results were within the same margin of at least 1 of the expert readers, indicating partial concordance. Conclusion: The AI-based method achieved high concordance with expert-derived TMTVs in a standardized benchmark setting. The findings demonstrate that the AI model performs comparably to human experts in most cases, even in an externally curated dataset deliberately enriched with challenging cases by its original authors. The AI model's ability to produce accurate, reproducible segmentations without user interaction could significantly reduce manual workload and interreader variability in lymphoma imaging. However, human supervision is required to minimize errors.</p>}},
  author       = {{Sadik, May and Larsson, Måns and Enqvist, Olof and Edenbrandt, Lars and Trägårdh, Elin}},
  issn         = {{0161-5505}},
  keywords     = {{artificial intelligence; benchmark dataset; deep learning; lymphoma; total metabolic tumor volume}},
  language     = {{eng}},
  number       = {{6}},
  pages        = {{1001--1005}},
  publisher    = {{Society of Nuclear Medicine Inc.}},
  series       = {{Journal of nuclear medicine : official publication, Society of Nuclear Medicine}},
  title        = {{Validation of an AI Method for Automated Lymphoma Metabolic Tumor Volume Segmentation Using a Public Benchmark PET/CT Dataset}},
  url          = {{http://dx.doi.org/10.2967/jnumed.125.271605}},
  doi          = {{10.2967/jnumed.125.271605}},
  volume       = {{67}},
  year         = {{2026}},
}