Validation of different automated segmentation models for target volume contouring in postoperative radiotherapy for breast cancer and regional nodal irradiation

• Verfasst von: Meixner, Eva [VerfasserIn]
• Verfasst von: Glogauer, Benjamin [VerfasserIn]
• Verfasst von: Klüter, Sebastian [VerfasserIn]
• Verfasst von: Wagner, Friedrich [VerfasserIn]
• Verfasst von: Neugebauer, David [VerfasserIn]
• Verfasst von: Hoeltgen, Line [VerfasserIn]
• Verfasst von: Dinges, Lisa A. [VerfasserIn]
• Verfasst von: Harrabi, Semi B. [VerfasserIn]
• Verfasst von: Liermann, Jakob [VerfasserIn]
• Verfasst von: Vinsensia, Maria [VerfasserIn]
• Verfasst von: Weykamp, Fabian [VerfasserIn]
• Verfasst von: Hoegen-Saßmannshausen, Philipp [VerfasserIn]
• Verfasst von: Debus, Jürgen [VerfasserIn]
• Verfasst von: Hörner-Rieber, Juliane [VerfasserIn]
• Titel: Validation of different automated segmentation models for target volume contouring in postoperative radiotherapy for breast cancer and regional nodal irradiation
• Verf.angabe: Eva Meixner, Benjamin Glogauer, Sebastian Klüter, Friedrich Wagner, David Neugebauer, Line Hoeltgen, Lisa A. Dinges, Semi Harrabi, Jakob Liermann, Maria Vinsensia, Fabian Weykamp, Philipp Hoegen-Saßmannshausen, Jürgen Debus, Juliane Hörner-Rieber
• E-Jahr: 2024
• Jahr: 11 September 2024
• Umfang: 7 S.
• Fussnoten: Gesehen am 04.06.2025
• Titel Quelle: Enthalten in: Clinical and translational radiation oncology
• Ort Quelle: Amsterdam : Elsevier, 2016
• Jahr Quelle: 2024
• Band/Heft Quelle: 49(2024), Artikel-ID 100855, Seite 100855-1-100855-7
• ISSN Quelle: 2405-6308
• DOI: doi:10.1016/j.ctro.2024.100855
• Datenträger: Online-Ressource
• Sprache: eng
• Sach-SW: AI contouring
• Sach-SW: Artificial intelligence
• Sach-SW: Auto-segmentation
• Sach-SW: Clinical implementation
• Sach-SW: Deep learning segmentation
• Sach-SW: Machine learning
• Sach-SW: Quality assurance
• Sach-SW: Target volume delineation
• K10plus-PPN: 1927457491

Abstract

Introduction - Target volume delineation is routinely performed in postoperative radiotherapy (RT) for breast cancer patients, but it is a time-consuming process. The aim of the present study was to validate the quality, clinical usability and institutional-specific implementation of different auto-segmentation tools into clinical routine. - Methods - Three different commercially available, artificial intelligence-, ESTRO-guideline-based segmentation models (M1-3) were applied to fifty consecutive reference patients who received postoperative local RT including regional nodal irradiation for breast cancer for the delineation of clinical target volumes: the residual breast, implant or chestwall, axilla levels 1 and 2, the infra- and supraclavicular regions, the interpectoral and internal mammary nodes. Objective evaluation metrics of the created structures were conducted with the Dice similarity index (DICE) and the Hausdorff distance, and a manual evaluation of usability. - Results - The resulting geometries of the segmentation models were compared to the reference volumes for each patient and required no or only minor corrections in 72 % (M1), 64 % (M2) and 78 % (M3) of the cases. The median DICE and Hausdorff values for the resulting planning target volumes were 0.87-0.88 and 2.96-3.55, respectively. Clinical usability was significantly correlated with the DICE index, with calculated cut-off values used to define no or minor adjustments of 0.82-0.86. Right or left sided target and breathing method (deep inspiration breath hold vs. free breathing) did not impact the quality of the resulting structures. - Conclusion - Artificial intelligence-based auto-segmentation programs showed high-quality accuracy and provided standardization and efficient support for guideline-based target volume contouring as a precondition for fully automated workflows in radiotherapy treatment planning.