End-to-End PET/CT Interpretation and Quantification with an LLM-Orchestrated AI Agent: A Real-World Pilot Study.
An AI system successfully performed complete PET scan interpretation and reporting autonomously, potentially freeing radiologists to focus on complex cases.
This pilot study in the Journal of Nuclear Medicine demonstrates that a large language model-orchestrated AI agent can autonomously perform the full PET/CT interpretation workflow—including lesion detection, quantification, and reporting—in real-world patients. The agentic AI paradigm applied to complex nuclear medicine imaging represents a significant methodological step toward clinical AI deployment.
What the study was
- Study design
- Prospective real-world pilot study
- Population
- Patients undergoing clinical PET/CT imaging
- Category
- Diagnostics
- Maturity
- Exploratory
- Journal
- Journal of Nuclear Medicine
Why it surfaced
LLM-orchestrated agentic AI for end-to-end PET/CT is a notable methodological novelty published in a leading nuclear medicine journal; pilot design and small team limit score, but the paradigm is high-signal for the AI diagnostics watchlist.
A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.