Multi-Modality Artificial Intelligence for Involved-Site Radiation Therapy: Clinical Target Volume Delineation in High-Risk Pediatric Hodgkin Lymphoma
Authored by: Tie, X., Milgrom, S.A., Lo, A.C., Charpentier, A.-M., LaRiviere, M.J., Maqbool, D., Cho, S.Y., Kelly, K.M., Hodgson, D., Castellino, S.M., Hoppe, B.S., Bradshaw, T.J.
📄 Related Publication:
Multi-Modality Artificial Intelligence for Involved-Site Radiation Therapy: Clinical Target Volume Delineation in High-Risk Pediatric Hodgkin Lymphoma
International Journal of Radiation Oncology · Biology · Physics (Red Journal)
Model Overview
This repository hosts deep learning models developed for automated clinical target volume (CTV) delineation in involved-site radiation therapy (ISRT) for high-risk pediatric Hodgkin lymphoma.
All models were trained and evaluated using imaging data from the Children’s Oncology Group (COG) AHOD1331 phase III clinical trial, a large, multi-institutional pediatric lymphoma dataset. The models are designed to integrate longitudinal, multi-modality imaging (i.e., baseline and interim PET/CT and planning CT images) to predict CTVs for radiation treatment planning.
Input Modalities
Depending on the model variant, inputs may include:
- Post-Chemotherapy Planning CT
- Baseline PET/CT (PET1)
- Interim PET/CT (PET2) (after 2 cycles of chemotherapy)
All PET/CT images are co-registered to the planning CT using either rigid or deformable registration, depending on the model configuration.
Available Model Variants
1. CT-only Models
- CT_only
- Input: Planning CT only
- Purpose: Baseline comparison against multi-modality approaches
2. Multi-Modality Early Fusion Models
- Early_fusion
- Inputs: Planning CT + baseline PET/CT + interim PET/CT
- Fusion strategy: Early fusion (channel-wise concatenation at input)
- Registration: Deformable registration for all modalities
3. Multi-Modality Late Fusion Models
- Late_fusion
- Inputs: Planning CT + baseline PET/CT + interim PET/CT
- Fusion strategy: Late fusion using architecture-specific feature integration
- Registration: Deformable registration for all modalities
Note that each variant has three models for ensemble.
4. Ablation Study Models (SwinUNETR)
Additional SwinUNETR models trained as part of ablation experiments are provided to assess the impact of imaging inputs and registration strategies:
PET_1_2_rigid
- Inputs: Planning CT + baseline PET/CT + interim PET/CT
- Registration: Rigid registration
PET_1_deform
- Inputs: Planning CT + baseline PET/CT (no interim PET/CT)
- Registration: Deformable registration
PET_1_rigid
- Inputs: Planning CT + baseline PET/CT (no interim PET/CT)
- Registration: Rigid registration
Each ablation folder contains both early-fusion and late-fusion SwinUNETR model weights.
Intended Use
These models are intended for research use only.
They are designed to serve as automated initial CTV contours to support ISRT planning workflows and must be reviewed and edited by radiation oncologists prior to any clinical application.
The models are not approved for clinical decision-making and have not undergone regulatory clearance.
Additional Resources
- Codebase (training, inference, evaluation):
https://github.com/xtie97/ISRT-CTV-AutoSeg
Citation
If you use these models in your research, please cite the associated publication:
@article{TIE2025,
title = {Multi-Modality Artificial Intelligence for Involved-Site Radiation Therapy: Clinical Target Volume Delineation in High-Risk Pediatric Hodgkin Lymphoma},
journal = {International Journal of Radiation Oncology*Biology*Physics},
year = {2025},
issn = {0360-3016},
doi = {https://doi.org/10.1016/j.ijrobp.2025.12.005},
url = {https://www.sciencedirect.com/science/article/pii/S0360301625065927},
author = {Xin Tie and Sarah A. Milgrom and Andrea C. Lo and Anne-Marie Charpentier and Michael J. LaRiviere and Danyal Maqbool and Steve Y. Cho and Kara M Kelly and David Hodgson and Sharon M. Castellino and Bradford S. Hoppe and Tyler J. Bradshaw}