Multicentre evaluation of deep learning CT autosegmentation of the head and neck region for radiotherapy

被引：0

作者：

Eric Pei Ping Pang ^{[1
]}

Hong Qi Tan ^{[2
]}

Fuqiang Wang ^{[1
]}

Jarkko Niemelä ^{[2
]}

Gregory Bolard ^{[3
]}

Susan Ramadan ^{[1
]}

Timo Kiljunen ^{[2
]}

Marta Capala ^{[4
]}

Steven Petit ^{[4
]}

Jan Seppälä ^{[5
]}

Kristiina Vuolukka ^{[5
]}

Ingrid Kiitam ^{[6
]}

Danil Zolotuhhin ^{[6
]}

Eduard Gershkevitsh ^{[7
]}

Kaisa Lehtiö ^{[7
]}

Juha Nikkinen ^{[8
]}

Jani Keyriläinen ^{[8
]}

Miia Mokka ^{[8
]}

Melvin Lee Kiang Chua ^{[9
]}

机构：

[1] National Cancer Centre Singapore,Division of Radiation Oncology

[2] Duke-NUS Graduate Medical School,Oncology Academic Clinical Programme

[3] Nanyang Technological University,Division of Physics and Applied Physics, School of Physical and Mathematical Science

[4] c/o Terkko Health hub,MVision Ai

[5] Docrates Cancer Center,Department of Radiotherapy, Erasmus MC Cancer Institute

[6] University Medical Center Rotterdam,Kuopio University Hospital

[7] Center of Oncology,Oulu University Hospital

[8] North Estonia Medical Centre,Research Unit of Health Sciences and Technology, Faculty of Medicine

[9] Department of Oncology and Radiotherapy,Turku University Hospital

[10] University of Oulu,Turku University Hospital

[11] Department of Oncology and Radiotherapy,University of Helsinki, Faculty of Science

[12] Department of Medical Physics,Division of Medical Sciences

[13] Department of Physics University of Helsinki,undefined

[14] National Cancer Centre Singapore,undefined

来源：

npj Digital Medicine | / 8卷 / 1期

关键词：

D O I：

10.1038/s41746-025-01624-z

中图分类号：

学科分类号：

摘要：

This is a multi-institutional study to evaluate a head-and-neck CT auto-segmentation software across seven institutions globally. 11 lymph node levels and 7 organs-at-risk contours were evaluated in a two-phase study design. Time savings were measured in both phases, and the inter-observer variability across the seven institutions was quantified in phase two. Overall time savings were found to be 42% in phase one and 49% in phase two. Lymph node levels IA, IB, III, IVA, and IVB showed no significant time savings, with some centers reporting longer editing times than manual delineation. All the edited ROIs showed reduced inter-observer variability compared to manual segmentation. Our study shows that auto-segmentation plays a crucial role in harmonizing contouring practices globally. However, the clinical benefits of auto-segmentation software vary significantly across ROIs and between clinics. To maximize its potential, institution-specific commissioning is required to optimize the clinical benefits.

引用