Clinical Implementation and Evaluation of Auto-Segmentation Tools for Multi-Site Contouring in Radiotherapy.

Auto-segmentation Deep Learning Radiotherapy Segmentation

Journal

Physics and imaging in radiation oncology
ISSN: 2405-6316
Titre abrégé: Phys Imaging Radiat Oncol
Pays: Netherlands
ID NLM: 101704276

Informations de publication

Date de publication:
Oct 2023
Historique:
received: 19 07 2023
revised: 09 11 2023
accepted: 09 11 2023
medline: 19 12 2023
pubmed: 19 12 2023
entrez: 19 12 2023
Statut: epublish

Résumé

Tools for auto-segmentation in radiotherapy are widely available, but guidelines for clinical implementation are missing. The goal was to develop a workflow for performance evaluation of three commercial auto-segmentation tools to select one candidate for clinical implementation. One hundred patients with six treatment sites (brain, head-and-neck, thorax, abdomen, and pelvis) were included. Three sets of AI-based contours for organs-at-risk (OAR) generated by three software tools and manually drawn expert contours were blindly rated for contouring accuracy. The dice similarity coefficient (DSC), the Hausdorff distance, and a dose/volume evaluation based on the recalculation of the original treatment plan were assessed. Statistically significant differences were tested using the Kruskal-Wallis test and the post-hoc Dunn Test with Bonferroni correction. The mean DSC scores compared to expert contours for all OARs combined were 0.80 ± 0.10, 0.75 ± 0.10, and 0.74 ± 0.11 for the three software tools. Physicians' rating identified equivalent or superior performance of some AI-based contours in head (eye, lens, optic nerve, brain, chiasm), thorax (e.g., heart and lungs), and pelvis and abdomen (e.g., kidney, femoral head) compared to manual contours. For some OARs, the AI models provided results requiring only minor corrections. Bowel-bag and stomach were not fit for direct use. During the interdisciplinary discussion, the physicians' rating was considered the most relevant. A comprehensive method for evaluation and clinical implementation of commercially available auto-segmentation software was developed. The in-depth analysis yielded clear instructions for clinical use within the radiotherapy department.

Sections du résumé

Background and purpose UNASSIGNED
Tools for auto-segmentation in radiotherapy are widely available, but guidelines for clinical implementation are missing. The goal was to develop a workflow for performance evaluation of three commercial auto-segmentation tools to select one candidate for clinical implementation.
Materials and Methods UNASSIGNED
One hundred patients with six treatment sites (brain, head-and-neck, thorax, abdomen, and pelvis) were included. Three sets of AI-based contours for organs-at-risk (OAR) generated by three software tools and manually drawn expert contours were blindly rated for contouring accuracy. The dice similarity coefficient (DSC), the Hausdorff distance, and a dose/volume evaluation based on the recalculation of the original treatment plan were assessed. Statistically significant differences were tested using the Kruskal-Wallis test and the post-hoc Dunn Test with Bonferroni correction.
Results UNASSIGNED
The mean DSC scores compared to expert contours for all OARs combined were 0.80 ± 0.10, 0.75 ± 0.10, and 0.74 ± 0.11 for the three software tools. Physicians' rating identified equivalent or superior performance of some AI-based contours in head (eye, lens, optic nerve, brain, chiasm), thorax (e.g., heart and lungs), and pelvis and abdomen (e.g., kidney, femoral head) compared to manual contours. For some OARs, the AI models provided results requiring only minor corrections. Bowel-bag and stomach were not fit for direct use. During the interdisciplinary discussion, the physicians' rating was considered the most relevant.
Conclusion UNASSIGNED
A comprehensive method for evaluation and clinical implementation of commercially available auto-segmentation software was developed. The in-depth analysis yielded clear instructions for clinical use within the radiotherapy department.

Identifiants

pubmed: 38111502
doi: 10.1016/j.phro.2023.100515
pii: S2405-6316(23)00106-9
pmc: PMC10726238
doi:

Types de publication

Journal Article

Langues

eng

Pagination

100515

Informations de copyright

© 2023 The Author(s).

Déclaration de conflit d'intérêts

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Auteurs

Gerd Heilemann (G)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Martin Buschmann (M)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Wolfgang Lechner (W)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Vincent Dick (V)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Franziska Eckert (F)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Martin Heilmann (M)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Harald Herrmann (H)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Matthias Moll (M)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Johannes Knoth (J)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Stefan Konrad (S)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Inga-Malin Simek (IM)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Christopher Thiele (C)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Alexandru Zaharie (A)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Dietmar Georg (D)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Joachim Widder (J)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Petra Trnkova (P)

Department of Radiation Oncology, Comprehensive Cancer Center Vienna, Medical University Vienna, Vienna, Austria.

Classifications MeSH