# Automated radiotherapy treatment planning guided by GPT-4Vision

arXiv: https://arxiv.org/abs/2406.15609 · HTML: https://mlsanity.com/paper/2406.15609/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 42 references that are in the map.
- First posted on arXiv 21 Jun 2024.

## Builds on (most cited first, 10 of 10)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Optuna: A Next-generation Hyperparameter Optimization Framework](https://mlsanity.com/paper/1907.10902/) (2019)
- [Reinforcement learning](https://mlsanity.com/paper/2405.10369/) (2024)
- [In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering](https://mlsanity.com/paper/2311.06668/) (2023)
- [Calibrating LLM-Based Evaluator](https://mlsanity.com/paper/2309.13308/) (2023)
- [Avoiding spurious correlations via logit correction](https://mlsanity.com/paper/2212.01433/) (2022)
- [RadOnc-GPT: A Large Language Model for Radiation Oncology](https://mlsanity.com/paper/2309.10160/) (2023)
- [Deep Probability Estimation](https://mlsanity.com/paper/2111.10734/) (2021)
- [Three-Dimensional Dose Prediction for Lung IMRT Patients with Deep Neural Networks: Robust Learning from Heterogeneous Beam Configurations](https://mlsanity.com/paper/1812.06934/) (2018)
- [Learning Image Representations for Content Based Image Retrieval of Radiotherapy Treatment Plans](https://mlsanity.com/paper/2206.02912/) (2022)

## Built on by (most cited first, 12 of 12)

- [Reducing Hallucinations in Vision-Language Models via Latent Space Steering](https://mlsanity.com/paper/2410.15778/) (2024)
- [Medical Reasoning in the Era of LLMs: A Systematic Review of Enhancement Techniques and Applications](https://mlsanity.com/paper/2508.00669/) (2025)
- [Enabling Scalable Evaluation of Bias Patterns in Medical LLMs](https://mlsanity.com/paper/2410.14763/) (2024)
- [MedPlan: A Two-Stage RAG-Based System for Personalized Medical Plan Generation](https://mlsanity.com/paper/2503.17900/) (2025)
- [Autonomous Radiotherapy Treatment Planning Using DOLA: A Privacy-Preserving, LLM-Based Optimization Agent](https://mlsanity.com/paper/2503.17553/) (2025)
- [Zero-Shot Large Language Model Agents for Fully Automated Radiotherapy Treatment Planning](https://mlsanity.com/paper/2510.11754/) (2025)
- [Benchmarking GPT-5 in Radiation Oncology: Measurable Gains, but Persistent Need for Expert Oversight](https://mlsanity.com/paper/2508.21777/) (2025)
- [An Iterative LLM Framework for SIBT utilizing RAG-based Adaptive Weight Optimization](https://mlsanity.com/paper/2509.08407/) (2025)
- [An Automated Retrieval-Augmented Generation LLaMA-4 109B-based System for Evaluating Radiotherapy Treatment Plans](https://mlsanity.com/paper/2509.20707/) (2025)
- [Towards Human-Centric Intelligent Treatment Planning for Radiation Therapy](https://mlsanity.com/paper/2510.13062/) (2025)
- [A Machine-to-Machine Knowledge-Guided LLM Agent for Generalizable Radiotherapy Treatment Planning](https://mlsanity.com/paper/2606.00922/) (2026)
- [CAREAgent: Clinical Agent with Structured Reasoning and Tool-Integrated for Order Generation](https://mlsanity.com/paper/2606.01094/) (2026)

## Abstract (arXiv)

Objective: Radiotherapy treatment planning is a time-consuming and potentially subjective process that requires the iterative adjustment of model parameters to balance multiple conflicting objectives. Recent advancements in frontier Artificial Intelligence (AI) models offer promising avenues for addressing the challenges in planning and clinical decision-making. This study introduces GPT-RadPlan, an automated treatment planning framework that integrates radiation oncology knowledge with the reasoning capabilities of large multi-modal models, such as GPT-4Vision (GPT-4V) from OpenAI. Approach: Via in-context learning, we incorporate clinical requirements and a few (3 in our experiments) approved clinical plans with their optimization settings, enabling GPT-4V to acquire treatment planning domain knowledge. The resulting GPT-RadPlan system is integrated into our in-house inverse treatment planning system through an application programming interface (API). For a given patient, GPT-RadPlan acts as both plan evaluator and planner, first assessing dose distributions and dose-volume histograms (DVHs), and then providing textual feedback on how to improve the plan to match the physician's requirements. In this manner, GPT-RadPlan iteratively refines the plan by adjusting planning parameters, such as weights and dose objectives, based on its suggestions. Main results: The efficacy of the automated planning system is showcased across 17 prostate cancer and 13 head and neck cancer VMAT plans with prescribed doses of 70.2 Gy and 72 Gy, respectively, where we compared GPT-RadPlan results to clinical plans produced by human experts. In all cases, GPT-RadPlan either outperformed or matched the clinical plans, demonstrating superior target coverage and reducing organ-at-risk doses by 5 Gy on average (15 percent for prostate and 10-15 percent for head and neck).