A feasibility study of automating radiotherapy planning with large language model agents
The DOI matches its Crossref record; 50 of the 53 references it cites were found in a registry or an index.
Abstract
Objective. Radiotherapy planning requires significant expertise to balance tumor control and organ-at-risk (OAR) sparing. Automated planning can improve both efficiency and quality. This study introduces GPT-Plan, a novel multi-agent system powered by the GPT-4 family of large language models (LLMs), for automating the iterative radiotherapy plan optimization. Approach. GPT-Plan uses LLM-driven agents, mimicking the collaborative clinical workflow of a dosimetrist and physicist, to iteratively generate and evaluate text-based radiotherapy plans based on predefined criteria. Supporting tools assist the agents by leveraging historical plans, mitigating LLM hallucinations, and balancing exploration and exploitation. Performance was evaluated on 12 lung (IMRT) and 5 cervical (VMAT) cancer cases, benchmarked against the ECHO auto-planning method and manual plans. The impact of historical plan retrieval on efficiency was also assessed. Results. For IMRT lung cancer cases, GPT-Plan generated high-quality plans, demonstrating superior target coverage and homogeneity compared to ECHO while maintaining comparable or better OAR sparing. For VMAT cervical cancer cases, plan quality was comparable to a senior physicist and consistently superior to a junior physicist, particularly for OAR sparing. Retrieving historical plans significantly reduced the number of required optimization iterations for lung cases (p < 0.01) and yielded iteration counts comparable to those of the senior physicist for cervical cases (p = 0.313). Occasional LLM hallucinations have been mitigated by self-reflection mechanisms. One limitation was the inaccuracy of vision-based LLMs in interpreting dose images. Significance. This pioneering study demonstrates the feasibility of automating radiotherapy planning using LLM-powered agents for complex treatment decision-making tasks. While challenges remain in addressing LLM limitations, ongoing advancements hold potential for further refining and expanding GPT-Plan’s capabilities.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Chain-Of-Thought Prompting Elicits Reasoning in Large Language Models
Wei, J., Wang, X., Schuurmans, D., Bosma, M., Ichter, B., Xia, F., Chi, E., Le, Q. V., & Zhou, D. (2022) · Advances in Neural Information Processing Systems 35 · doi:10.52202/068431-1800 · Cited by 4,505
MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework
Hong, S., Zhuge, M., Chen, J., Zheng, X., Cheng, Y., Zhang, C., Wang, J., Wang, Z., Yau, S. K. S., Lin, Z., et al. (2023) · arXiv · doi:10.48550/arxiv.2308.00352 · Cited by 2,825 · cited 1×
“…capable of reasoning and decision-making (Bubeck et al 2023, Yang et al 2023) and have been successfully integrated into multi-agent systems, effectively mimicking the collaborative nature of human experts to solve complex problems (Hong et al 2023, Park et al 2023, Xu et al 2023, Guo et al 2024).”
Self-citationGenerative Agents: Interactive Simulacra of Human Behavior
Park, J. S., O'Brien, J., Cai, C. J., Morris, M. R., Liang, P., & Bernstein, M. S. (2023) · Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology · doi:10.1145/3586183.3606763 · Cited by 2,187 · cited 1×
“…capable of reasoning and decision-making (Bubeck et al 2023, Yang et al 2023) and have been successfully integrated into multi-agent systems, effectively mimicking the collaborative nature of human experts to solve complex problems (Hong et al 2023, Park et al 2023, Xu et al 2023, Guo et al 2024).”
Generative Agents: Interactive Simulacra of Human Behavior
Park, J. S., O'Brien, J. C., Cai, C. J., Morris, M. R., Liang, P., & Bernstein, M. S. (2023) · arXiv · doi:10.48550/arxiv.2304.03442 · Cited by 2,173 · cited 1×
“…capable of reasoning and decision-making (Bubeck et al 2023, Yang et al 2023) and have been successfully integrated into multi-agent systems, effectively mimicking the collaborative nature of human experts to solve complex problems (Hong et al 2023, Park et al 2023, Xu et al 2023, Guo et al 2024).”
Foundation models for generalist medical artificial intelligence
Moor, M., Banerjee, O., Abad, Z. S. H., Krumholz, H. M., Leskovec, J., Topol, E. J., & Rajpurkar, P. (2023) · Nature · doi:10.1038/s41586-023-05881-4 · Cited by 2,028 · cited 1×
“Despite these current limitations, future advancements and fine-tuning with diverse medical images, including dose distributions, CT/MR scans (Moor et al 2023), and even pathology slides (Lu et al 2024), hold promise for enhancing vLLM accuracy.”
Large Language Model based Multi-Agents: A Survey of Progress and Challenges
Guo, T., Chen, X., Wang, Y., Chang, R., Pei, S., Chawla, N. V., Wiest, O., & Zhang, X. (2024) · arXiv · doi:10.48550/arxiv.2402.01680 · Cited by 1,713
Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., Küttler, H., Lewis, M., Yih, W.-T., Rocktäschel, T., et al. (2020) · UCL Discovery (University College London) · Cited by 1,187
Reflexion: Language Agents with Verbal Reinforcement Learning
Shinn, N., Cassano, F., Berman, E., Gopinath, A., Narasimhan, K., & Yao, S. (2023) · arXiv · doi:10.48550/arxiv.2303.11366 · Cited by 1,087 · cited 1×
“To mitigate this, we employ a two-stage self-reflection mechanism (Shinn et al 2024) as outlined in figure 3.”
Vision-Language Models for Vision Tasks: A Survey
Zhang, J., Huang, J., Jin, S., & Lu, S. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3369699 · Cited by 1,051 · cited 2×
“Furthermore, current vision LLMs (vLLMs) (Zhang et al 2024) demonstrate limitations in accurately interpreting dose distributions images, primarily due to insufficient training on radiotherapy-specific data.”
Gemini: A Family of Highly Capable Multimodal Models
Gemini Team, Anil, R., Borgeaud, S., Alayrac, J.-B., Yu, J., Soricut, R., Schalkwyk, J., Dai, A. M., Hauth, A., Millican, K., et al. (2023) · arXiv · doi:10.48550/arxiv.2312.11805 · Cited by 990
Research works printed but not linked: 1
- Yang, 2023, The dawn of LMMs: preliminary explorations with GPT-4V(ision), vol 9, p 1
Cited by
The role of agentic artificial intelligence in healthcare: a scoping review
Collaço, B. G., Haider, S. A., Prabha, S., Gomez-Cabello, C. A., Genovese, A., Wood, N. G., Bagaria, S. P., Gopala, N., Tao, C., & Forte, A. J. (2026) · npj Digital Medicine · doi:10.1038/s41746-026-02517-5 · Cited by 127
“Radiation oncology has emerged as a particularly promising fi eld for Agentic AI development, as demonstrated by Gu et al. (2025) 16 , Wang et al. (2025) 14 , and Yang et al. (2025) 17 , who introduced experimental systems for diagnostic reasoning, treatment planning, and cancer detection.”
AI Agents in Clinical Medicine: A Systematic ReviewPreprint
Gorenshtein, A., Omar, M., Glicksberg, B. S., Nadkarni, G. N., & Klang, E. (2025) · medRxiv · doi:10.1101/2025.08.22.25334232 · Cited by 94
Automated radiotherapy treatment planning guided by GPT-4Vision
Liu, S., Pastor‐Serrano, O., Chen, Y., Gopaulchan, M., Liang, W., Buyyounouski, M. K., Pollom, E. L., Le, Q., Gensheimer, M. F., Dong, P., et al. (2025) · Physics in Medicine and Biology · doi:10.1088/1361-6560/adf02c · Cited by 56
“A recent study has also explored a similar agentic framework for treatment planning (Wang et al 2025), further highlighting the growing importance and potential of LLM-based approaches in radiation oncology.”
A comprehensive survey of AI agents in healthcare
Xu, G., Li, X., Chen, Y., Duan, Y., Wu, S., Yu, H., Chiu, C.-H., Ni, J., Tang, N., Li, T. J.-J., et al. (2026) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2026.105045 · Cited by 56
Artificial intelligence agents in healthcare research: A scoping review
Njei, B., Al-Ajlouni, Y. A., Sidney Kanmounye, U., Boateng, S., Loic Nguefang, G., Njei, N., Hamouri, S., & Al-Ajlouni, A. F. (2026) · PLOS One · doi:10.1371/journal.pone.0342182 · Cited by 27
“Cloud-based LLMs like GPT-4 are often considered unsuitable for handling sensitive patient data due to regulatory restrictions [17,31], necessitating the exploration of locally deployed, open-source LLMs to mitigate data transfer risks [20].”
Influential
Metadata
- Authors
- Wang, Q., Wang, Z., Li, M., Ni, X., Tan, R., Zhang, W., Wubulaishan, M., Wang, W., Yuan, Z., Zhang, Z., & Liu, C.
- Affiliations
- Tianjin University (CN)
- Tianjin Medical University Cancer Institute and Hospital (CN)
- Maastricht University Medical Centre (NL)
- Changzhou No.2 People's Hospital (CN)
- Shanghai Business School (CN)
- Chinese Academy of Sciences (CN)
- ORCID
- Venue
- Physics in Medicine & Biology
- Publisher
- IOP Publishing
- Published
- 2025-03-21
- Volume
- 70
- Issue
- 7
- Pages
- 075007
- ISSN
- 0031-9155 (ISSN-L)
- 1361-6560
- OpenAlex
- W4408388242
- Cited by
- 28
- References
- 53
- Citations by year
- 2026: 21 · 2025: 7
- FWCI
- 17.64
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- iopscience.iop.org/page/copyright (vor)
- iopscience.iop.org/info/page/text-and-data-mining (tdm)
- Locations
- Physics in Medicine and Biology (publishedVersion)
- PubMed (publishedVersion)
- Research Publications (Maastricht University) (submittedVersion)
- Funders
- National Natural Science Foundation of China (62371243)
- Tianjin Key Medical Discipline (Specialty) Construction Project (TJYXZDXK-009A)
- Topics
- Physical Sciences › Physics and Astronomy › Radiation › Advanced Radiotherapy Techniques
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Physical Sciences › Physics and Astronomy › Radiation › Radiation Therapy and Dosimetry
- Keywords
- large language models, multi-agent systems, GPT-4, organs at risk sparing, lung cancer, cervical cancer, intensity-modulated radiotherapy, volumetric modulated arc therapy, LLM hallucination mitigation, plan optimization
- MeSH terms
- Large Language Models, Automation, Uterine Cervical Neoplasms, Feasibility Studies, Female, Humans, Lung Neoplasms, Programming Languages, Radiotherapy Dosage, Radiotherapy Planning, Computer-Assisted, Radiotherapy, Intensity-Modulated, Organs at Risk
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: A feasibility study of automating radiotherapy planning with large language model agents. https://app.cochat.ai/papers/10.1088/1361-6560/adbff1
- The paper
- Wang, Q., Wang, Z., Li, M., Ni, X., Tan, R., Zhang, W., Wubulaishan, M., Wang, W., Yuan, Z., Zhang, Z., & Liu, C. (2025). A feasibility study of automating radiotherapy planning with large language model agents. Physics in Medicine & Biology. https://doi.org/10.1088/1361-6560/adbff1
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.