RadOnc-Agent: An LLM-Orchestrated Framework for AI Workflows Across the Radiotherapy Care Pathway
Organizations: Department of Radiation Oncology, Mayo Clinic Arizona, Phoenix, AZ, USA.
Abstract
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.
Figures & tables
| Function group | Callable function | Function description | Headline validation result | Evidence status |
|---|---|---|---|---|
| Diagnosis & Decision | RT-indication assessment | Retrieves guideline-supported radiotherapy indications and fractionation recommendations with source citations. | Top-5 retrieval 97.50%; overall answer accuracy 95.33%; unsupported-query withholding 96.67%. | Current task-level evaluation |
| Modality selection | Compares candidate radiotherapy modalities using patient context and, when available, verified quantitative risk parameters. | Completely acceptable regimen 46–52%; acceptable including minor deviations 64–66%; modality-selection accuracy 70.5–72.2%. | Previously published task-level evaluation | |
| Simulation & Planning | Planning directive | Converts diagnosis, treatment intent and prescription into a structured planning directive. | Supporting order-summarization evaluation: more than 98% agreement with therapist-verified references. | Previously published task-level evaluation |
| MR-to-CT synthesis | Generates a planning-compatible synthetic CT from planning MRI. | MAE 46.8 12.7 HU; SSIM 0.961 0.018. | Completed internal task-level evaluation | |
| Deformable registration | Registers longitudinal CT images and returns a deformation field and propagated anatomy. | OAR/target Dice 92.07%/87.14%; OAR folding 0.08%; accumulated-dose MAE 0.72%. | Completed internal task-level evaluation | |
| Callable function | Development or training data | Evaluation data | Reference standard and endpoints |
|---|---|---|---|
| RT-indication assessment | BM25 retrieval plus general-language-model synthesis; no task-specific parameter training. | 300 prespecified questions across ten disease sites: 240 supported and 60 unsupported or out of corpus. | Radiation-oncologist-adjudicated answers, supporting passages and refusal status; retrieval, answer, citation and withholding endpoints. |
| Modality selection | Fine-tuned LLaMA2 and Mistral models; 7,500 curated development cases. | Independent test set of 403 patient cases [ 11 ] . | Radiation-oncologist assessment of regimen acceptability and agreement with the documented treatment modality. |
| Planning directive | General-language-model order summarization; no task-specific parameter training. | 607 CT-simulation orders from seven treatment-modality and disease-site groups [ 12 ] . | Therapist-verified order summaries; summary accuracy and consistency. |
| MR-to-CT synthesis | 2.5D U-Net++; public SynthRAD2025 release, with 462 training and 58 validation cases. | 58 patient-level held-out SynthRAD2025 MRI-to-CT cases [ 22 ] . | Reference CT; regional HU MAE and SSIM. |
| Deformable registration | Patient-adaptive registration network; 2,052 patients/7,429 image pairs for training and 293 patients/1,062 pairs for validation. | Internal test: 586 patients/2,122 image pairs; external test: 542 patients/1,373 pairs from two private and five public cohorts [ 13 ] . | Expert contours and reference dose; Dice, folding and accumulated-dose MAE. |
| Auto-contouring | Treatment-plan-conditioned three-dimensional segmentation network; 3,355 training and 387 validation plans. | Pooled two-centre held-out test set of 370 plans from 279 patients, plus fully held-out external cohorts [ 14 ] . | Physician-defined target and OAR contours; Dice and inference time. |
| Execution risk | Detection mechanism | Fail-safe behavior | Control mode |
|---|---|---|---|
| Missing required input | Function-specific schema checks identify absent required arguments before backend routing. | Backend dispatch is blocked; the agent issues a clarification or withholding response and can resume the workflow only after the required information is available. | Deterministic gate with LLM-mediated clarification |
| Unregistered function, malformed argument or invalid route | The function registry validates the function name, required and optional arguments, and registered backend route. | The call is rejected before backend invocation rather than routed to an unregistered or incompatible service. | Deterministic |
| Patient, course or plan identifier mismatch | Identifiers in each call are checked against the active session; structured outputs retain patient-, course- and plan-level provenance. | The inconsistent call is rejected and is not merged into the active patient/workflow context. | Deterministic |
| Cross-patient context carryover | Session state is bound to one active patient and treatment course, with no persistent conversational memory or cross-patient cache; state is cleared between independent executions. | Information from a different session is not retained as active patient or workflow context for subsequent calls. | Deterministic session isolation |
| Transient backend or asynchronous-job failure | Backend status, transport errors and asynchronous-job progress are monitored by the dispatch layer. | Workflow state is retained for bounded retry or resumption after service restoration; an unresolved failure is returned explicitly rather than treated as a successful result. | Hybrid deterministic handling and LLM-mediated resumption |
Workflow Automation
Workflow automation coordinates multi-step tasks across applications, tools, and people using software, AI agents, and explicit process logic.
See all 62 papers