Generative AI in Radiotherapy Imaging: From Cross-Modality Synthesis to Foundation Model-Based Segmentation.
Abstract
Imaging underpins nearly every stage of radiation therapy, yet the images obtained closest to treatment delivery are often the least reliable. Cone-beam CT is degraded by scatter artifacts and inaccurate Hounsfield unit values, whereas on-treatment MRI is limited by time constraints that preclude standard protocols. Generative artificial intelligence provides new solutions to these limitations. This review examines three areas of direct impact: cross-modality image synthesis, in which diffusion models generate CT-quality images foundation model-based segmentation and target delineation, where large-scale pretrained models generalize across modalities without task-specific retraining and multimodal architectures improve delineation by integrating imaging with clinical text; and generative approaches to post-treatment assessment. Persistent challenges, including image hallucinations, algorithmic biases, and data privacy, must be addressed before these tools can be deployed safely.
Related
- Contrastive adversarial mix-up learning for unsupervised cross-modality image segmentation
- Modality-mapping-guided customized CycleGAN filtering to enhance narrow-band imaging for improved colorectal lesion segmentation
- Automated thigh muscle segmentation using foundation models: validation and impact of segmentation quality
- Data-Efficient Prostate Micro-Ultrasound Image Analysis using a Modality-specific Foundation Model
- Foundation model-driven framework for human–object interaction prediction with segmentation mask integration
BibTeX
@article{park2026generative,
title = {{Generative AI in Radiotherapy Imaging: From Cross-Modality Synthesis to Foundation Model-Based Segmentation.}},
author = {Sangjoon Park},
journal = {Journal of the Korean Society of Radiology},
year = {2026},
month = sep,
doi = {10.3348/jksr.2026.0084},
url = {https://europepmc.org/article/MED/42903063},
note = {CC BY-NC},
}
RIS
TY - JOUR
TI - Generative AI in Radiotherapy Imaging: From Cross-Modality Synthesis to Foundation Model-Based Segmentation.
AU - Sangjoon Park
PY - 2026
DA - 2026/09/29
JO - Journal of the Korean Society of Radiology
DO - 10.3348/jksr.2026.0084
UR - https://europepmc.org/article/MED/42903063
AB - Imaging underpins nearly every stage of radiation therapy, yet the images obtained closest to treatment delivery are often the least reliable. Cone-beam CT is degraded by scatter artifacts and inaccurate Hounsfield unit values, whereas on-treatment MRI is limited by time constraints that preclude standard protocols. Generative artificial intelligence provides new solutions to these limitations. This review examines three areas of direct impact: cross-modality image synthesis, in which diffusion models generate CT-quality images foundation model-based segmentation and target delineation, where large-scale pretrained models generalize across modalities without task-specific retraining and multimodal architectures improve delineation by integrating imaging with clinical text; and generative approaches to post-treatment assessment. Persistent challenges, including image hallucinations, algorithmic biases, and data privacy, must be addressed before these tools can be deployed safely.
SN - 2951-0805
ER -