RadOnc-Agent: An LLM-Orchestrated Framework for AI Workflows Across the Radiotherapy Care Pathway
Quick Answer
RadOnc-Agent is an AI framework that integrates 26 functions across four radiotherapy phases, achieving 98.79% accuracy in function selection and 96.50% completion in scripted workflows.
Quick Take
The system demonstrates the potential of orchestration in unifying disparate radiotherapy tasks, although clinical correctness remains unverified.
Key Points
- RadOnc-Agent formalizes radiotherapy into four clinical phases.
- Achieved 96.67% completion rate in real-patient workflow executions.
- Removing longitudinal state decreased cross-stage completion from 96.50% to 84.00%.
- The framework preserves patient context and routes requests effectively.
- Findings support the feasibility of LLM orchestration in radiotherapy.
DeepSignal Analysis
What happened
RadOnc-Agent is an AI framework designed to streamline radiotherapy workflows by integrating 26 functions across four clinical phases. It achieved a function selection accuracy of 98.79% and a workflow completion rate of 96.50%. However, the framework's clinical correctness and utility have not been validated.
Key evidence
- RadOnc-Agent integrates 26 callable functions across four phases of radiotherapy, aiming to unify fragmented tasks.
- The framework achieved 98.79% accuracy in selecting intended functions and 96.50% completion in scripted workflows.
- Comparative evaluations showed that removing longitudinal state reduced cross-stage workflow completion from 96.50% to 84.00%.
Why it matters
The development of RadOnc-Agent highlights the potential for AI to enhance the efficiency of radiotherapy by coordinating various tasks and information. However, the lack of verification regarding clinical correctness raises concerns about its practical application in patient care. As AI continues to evolve in healthcare, understanding its limitations is crucial for safe implementation.
What to watch
Paper Resources
📖 Reader Mode
~2 min readAbstract:Artificial intelligence has advanced individual radiotherapy tasks, yet these capabilities remain separated across clinical stages, software environments and data modalities. This fragmentation contrasts with the longitudinal radiotherapy workflow from treatment decision-making through follow-up. Here we present RadOnc-Agent, an agentic artificial-intelligence framework that formalizes radiotherapy into four clinical phases and provides 26 callable functions through a conversational interface. A large-language-model controller maps clinical intent to schema-constrained calls, preserves patient and workflow context, and routes requests to specialist services. We evaluated system execution using 2,600 single-function requests (7,800 repeat executions), 200 prespecified synthetic cross-stage scenarios spanning four phases (600 executions), and 120 workflow instances from 60 de-identified patient records (360 clean executions) representing decision-to-planning and planning-to-adaptation. RadOnc-Agent selected the intended function in 98.79% of single-function executions, completed 96.50% of scripted cross-stage workflows, and completed 96.67% of real-patient workflow executions. In comparative ablations, removing longitudinal state reduced cross-stage completion from 96.50% to 84.00%, while disabling schema and identity validation increased mismatched backend dispatch from 0% to 95.28% in a replay/test evaluation. These findings establish the technical feasibility of an LLM-orchestrated architecture for coordinating heterogeneous radiotherapy capabilities and information across longitudinal workflows; they do not establish clinical correctness, clinical utility or prospective benefit.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.06923 [cs.AI] |
| (or arXiv:2610.06923v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.06923 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Caiwen Jiang [view email]
[v1]
Fri, 2 Oct 2026 22:39:48 UTC (29,623 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.AI
See more →HOBA: Hierarchical On-Policy Bidding Agents for Adaptive Online Advertising
HOBA (Hierarchical On-policy Bidding Agents) is a novel hierarchical reinforcement learning framework that enhances online advertising bidding systems by improving adaptability and reducing hyperparameter tuning costs. It utilizes a for hyperparameter inference, a SARSA agent for expert model selection, and a dynamic expert pool for bid execution, achieving a +3.6% increase in target cost during large-scale deployment and outperforming state-of-the-art baselines on AuctionNet.