RadOnc-GPT: An Autonomous LLM Agent for Real-Time Patient Outcomes Labeling at Scale
This addresses the need for scalable and timely patient outcomes labeling in radiation oncology, though it appears incremental as an application of existing LLM methods to a new medical domain.
The authors tackled the problem of manual labeling limiting patient outcomes research in radiation oncology by developing RadOnc-GPT, an autonomous LLM agent for real-time labeling, which achieved validation in structured QA and complex clinical outcomes tasks, including detecting mandibular osteoradionecrosis and cancer recurrence with specific accuracy metrics.
Manual labeling limits the scale, accuracy, and timeliness of patient outcomes research in radiation oncology. We present RadOnc-GPT, an autonomous large language model (LLM)-based agent capable of independently retrieving patient-specific information, iteratively assessing evidence, and returning structured outcomes. Our evaluation explicitly validates RadOnc-GPT across two clearly defined tiers of increasing complexity: (1) a structured quality assurance (QA) tier, assessing the accurate retrieval of demographic and radiotherapy treatment plan details, followed by (2) a complex clinical outcomes labeling tier involving determination of mandibular osteoradionecrosis (ORN) in head-and-neck cancer patients and detection of cancer recurrence in independent prostate and head-and-neck cancer cohorts requiring combined interpretation of structured and unstructured patient data. The QA tier establishes foundational trust in structured-data retrieval, a critical prerequisite for successful complex clinical outcome labeling.