Generating Synthetic Computed Tomography for Radiotherapy: SynthRAD2023 Challenge Report
Abstract
Radiation therapy plays a crucial role in cancer treatment, necessitating precise delivery of radiation to tumors while sparing healthy tissues over multiple days. Computed tomography (CT) is integral for treatment planning, offering electron density data crucial for accurate dose calculations. However, accurately representing patient anatomy is challenging, especially in adaptive radiotherapy, where CT is not acquired daily. Magnetic resonance imaging (MRI) provides superior soft-tissue contrast. Still, it lacks electron density information while cone beam CT (CBCT) lacks direct electron density calibration and is mainly used for patient positioning. Adopting MRI-only or CBCT-based adaptive radiotherapy eliminates the need for CT planning but presents challenges. Synthetic CT (sCT) generation techniques aim to address these challenges by using image synthesis to bridge the gap between MRI, CBCT, and CT. The SynthRAD2023 challenge was organized to compare synthetic CT generation methods using multi-center ground truth data from 1080 patients, divided into two tasks: 1) MRI-to-CT and 2) CBCT-to-CT. The evaluation included image similarity and dose-based metrics from proton and photon plans. The challenge attracted significant participation, with 617 registrations and 22/17 valid submissions for tasks 1/2. Top-performing teams achieved high structural similarity indices (>0.87/0.90) and gamma pass rates for photon (>98.1%/99.0%) and proton (>97.3%/97.0%) plans. However, no significant correlation was found between image similarity metrics and dose accuracy, emphasizing the need for dose evaluation when assessing the clinical applicability of sCT. SynthRAD2023 facilitated the investigation and benchmarking of sCT generation techniques, providing insights for developing MRI-only and CBCT-based adaptive radiotherapy.
Cite
@article{arxiv.2403.08447,
title = {Generating Synthetic Computed Tomography for Radiotherapy: SynthRAD2023 Challenge Report},
author = {Evi M. C. Huijben and Maarten L. Terpstra and Arthur Jr. Galapon and Suraj Pai and Adrian Thummerer and Peter Koopmans and Manya Afonso and Maureen van Eijnatten and Oliver Gurney-Champion and Zeli Chen and Yiwen Zhang and Kaiyi Zheng and Chuanpu Li and Haowen Pang and Chuyang Ye and Runqi Wang and Tao Song and Fuxin Fan and Jingna Qiu and Yixing Huang and Juhyung Ha and Jong Sung Park and Alexandra Alain-Beaudoin and Silvain Bériault and Pengxin Yu and Hongbin Guo and Zhanyao Huang and Gengwan Li and Xueru Zhang and Yubo Fan and Han Liu and Bowen Xin and Aaron Nicolson and Lujia Zhong and Zhiwei Deng and Gustav Müller-Franzes and Firas Khader and Xia Li and Ye Zhang and Cédric Hémon and Valentin Boussot and Zhihao Zhang and Long Wang and Lu Bai and Shaobin Wang and Derk Mus and Bram Kooiman and Chelsea A. H. Sargeant and Edward G. A. Henderson and Satoshi Kondo and Satoshi Kasai and Reza Karimzadeh and Bulat Ibragimov and Thomas Helfer and Jessica Dafflon and Zijie Chen and Enpei Wang and Zoltan Perko and Matteo Maspero},
journal= {arXiv preprint arXiv:2403.08447},
year = {2024}
}
Comments
Preprint submitted to Medical Image Analysis