학술논문
Comprehensive deep learning-based framework for automatic organs-at-risk segmentation in head-and-neck and pelvis for MR-guided radiation therapy planning
Document Type
article
Author
Vanda Czipczer; Bernadett Kolozsvári; Borbála Deák-Karancsi; Marta E. Capala; Rachel A. Pearson; Emőke Borzási; Zsófia Együd; Szilvia Gaál; Gyöngyi Kelemen; Renáta Kószó; Viktor Paczona; Zoltán Végváry; Zsófia Karancsi; Ádám Kékesi; Edina Czunyi; Blanka H. Irmai; Nóra G. Keresnyei; Petra Nagypál; Renáta Czabány; Bence Gyalai; Bulcsú P. Tass; Balázs Cziria; Cristina Cozzini; Lloyd Estkowsky; Lehel Ferenczi; András Frontó; Ross Maxwell; István Megyeri; Michael Mian; Tao Tan; Jonathan Wyatt; Florian Wiesinger; Katalin Hideghéty; Hazel McCallum; Steven F. Petit; László Ruskó
Source
Frontiers in Physics, Vol 11 (2023)
Subject
Language
English
ISSN
2296-424X
Abstract
Introduction: The excellent soft-tissue contrast of magnetic resonance imaging (MRI) is appealing for delineation of organs-at-risk (OARs) as it is required for radiation therapy planning (RTP). In the last decade there has been an increasing interest in using deep-learning (DL) techniques to shorten the labor-intensive manual work and increase reproducibility. This paper focuses on the automatic segmentation of 27 head-and-neck and 10 male pelvis OARs with deep-learning methods based on T2-weighted MR images.Method: The proposed method uses 2D U-Nets for localization and 3D U-Net for segmentation of the various structures. The models were trained using public and private datasets and evaluated on private datasets only.Results and discussion: Evaluation with ground-truth contours demonstrated that the proposed method can accurately segment the majority of OARs and indicated similar or superior performance to state-of-the-art models. Furthermore, the auto-contours were visually rated by clinicians using Likert score and on average, 81% of them was found clinically acceptable.