Interpretable graph‐conditioned CNNs for dose‐volume histogram prediction in radiotherapy
Medical Physics · 2026
DOI: 10.1002/mp.70663 ↗At a glance
Journal article in Medical Physics, 2026. Written by Xiaoling Zhu and 7 co-authors. Institution: Department of Radiation Oncology, Fudan University Shanghai Cancer Center, Shanghai, China. Major MEDLINE topics: Convolutional Neural Networks, Radiotherapy Planning, Computer-Assisted, Radiation Dosage. 32 references. The full text is freely available in PubMed Central. Funded by: Beijing Natural Science Foundation, Shanghai Pujiang Programme, National High Level Hospital Clinical Research Funding.
Abstract (original language)
Background
Accurate dose‐volume histogram (DVH) prediction is essential for high‐quality and automated radiotherapy (RT) planning. However, existing deep learning methods primarily focus on voxel‐level dose prediction followed by post‐processing to extract DVHs, a workflow that can introduce cumulative errors and limit clinical interpretability.
Purpose
In this study, we propose a novel deep learning framework that directly predicts full DVHs for multiple organs‐at‐risk (OARs) from computed tomography (CT) and structure contours.
Methods
Our method combines a three‐dimensional convolutional neural network (CNN) for anatomical feature extraction with a graph neural network (GNN) that models each DVH as a structured graph, enabling sequential dose‐volume dependencies to be learned explicitly. The model was extensively validated on nasopharyngeal cancer cases from an external institute and rectal cancer cases from our affiliated hospital, encompassing treatments with TomoTherapy and volumetric modulated arc therapy (VMAT). We additionally conducted a model‐guided clinical DVH assessment on suboptimal cases to demonstrate the method's clinical utility Results The proposed method achieved low dose‐wise prediction errors and outperformed previous approach. Specifically, it reduced the mean dose error for the brain stem from 1.24 Gy to 0.66 Gy and for the larynx from 1.53 Gy to 0.78 Gy, with notable improvements also observed in the parotid glands and temporal lobes. Statistical analysis of clinical indices demonstrated non‐inferiority to ground‐truth plans ( p > 0.05). Moreover, the predicted DVHs effectively guided plan revision and improved plan quality.
Conclusions
Our results suggest that direct DVH prediction using a CNN‐GNN framework offers a robust and clinically interpretable solution for treatment planning and quality assurance.
Subjects
MeSH terms (MEDLINE)
Author keywords
Authors and institutions
- Department of Radiation Oncology, Fudan University Shanghai Cancer Center, Shanghai, China.
- Department of Oncology, Shanghai Medical College, Fudan University, Shanghai, China.
- Shanghai Clinical Research Center for Radiation Oncology, Shanghai, China.
- Shanghai Key Laboratory of Radiation Oncology, Shanghai, China.
- Department of Radiation Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China.
Funding
- Beijing Natural Science Foundation
- Shanghai Pujiang Programme
- National High Level Hospital Clinical Research Funding
- National Natural Science Foundation of China
- Natural Science Foundation of Beijing Municipality
Cite
APA
Zhu, X., Ling, C., Xu, H., Men, K., Dai, J., Hu, W., Liu, Z., & Fan, J. (2026). Interpretable graph‐conditioned CNNs for dose‐volume histogram prediction in radiotherapy. Medical Physics. https://doi.org/10.1002/mp.70663
Vancouver
Zhu X, Ling C, Xu H, Men K, Dai J, Hu W, et al. Interpretable graph‐conditioned CNNs for dose‐volume histogram prediction in radiotherapy. Medical Physics. 2026. doi:10.1002/mp.70663
BibTeX
@article{Zhu2026,
author = {Zhu, Xiaoling and Ling, Chifang and Xu, Hao and Men, Kuo and Dai, Jianrong and Hu, Weigang and Liu, Zhiqiang and Fan, Jiawei},
title = {Interpretable graph‐conditioned CNNs for dose‐volume histogram prediction in radiotherapy},
journal = {Medical Physics},
year = {2026},
doi = {10.1002/mp.70663}
}
EUVIMED