Paper
Automated Extraction of Unstructured Post-SBRT Toxicity Data from Radiology Reports Using Large Language Models
Authors
Justin Pijanowski, Yakout Mezgueldi, Alan Lee, Drew Moghanaki, Ricky R. Savjani, James Lamb
Abstract
We evaluated the viability of using a Large Language Model (LLM) to extract patient-specific specific toxicity and progression outcomes from unstructured radiology reports. We retrospectively extracted 160 follow-up CT and PET/CT electronic medical record notes for patients treated with lung stereotactic body radiotherapy (SBRT) at our institution from January 2017 through December 2023. Using the Llama 3.3-70-B-Instruct LLM, we engineered prompts to extract four clinical endpoints from each radiology report: locoregional progression, distant progression, radiation-induced fibrosis, and radiation-induced rib fractures. Progression endpoints were classified as yes, no, or maybe, while fibrosis and rib fractures were binary (yes or no). Ground truth labels were defined using two-grader consensus for the 60-note training set, used for prompt development, and a three-grader majority vote for the 100-note test set. LLM performance was evaluated using sensitivity, specificity, and accuracy. As detailed by our evaluation metrics, the strong performance of our methods demonstrates the viability of using prompt-engineered LLMs to extract radiation-toxicities and progression classification from radiology reports.
Metadata
Related papers
Vibe Coding XR: Accelerating AI + XR Prototyping with XR Blocks and Gemini
Ruofei Du, Benjamin Hersh, David Li, Nels Numan, Xun Qian, Yanhe Chen, Zhongy... • 2026-03-25
Comparing Developer and LLM Biases in Code Evaluation
Aditya Mittal, Ryan Shar, Zichu Wu, Shyam Agarwal, Tongshuang Wu, Chris Donah... • 2026-03-25
The Stochastic Gap: A Markovian Framework for Pre-Deployment Reliability and Oversight-Cost Auditing in Agentic Artificial Intelligence
Biplab Pal, Santanu Bhattacharya • 2026-03-25
Retrieval Improvements Do Not Guarantee Better Answers: A Study of RAG for AI Policy QA
Saahil Mathur, Ryan David Rittner, Vedant Ajit Thakur, Daniel Stuart Schiff, ... • 2026-03-25
MARCH: Multi-Agent Reinforced Self-Check for LLM Hallucination
Zhuo Li, Yupeng Zhang, Pengyu Cheng, Jiajun Song, Mengyu Zhou, Hao Li, Shujie... • 2026-03-25
Raw Data (Debug)
{
"raw_xml": "<entry>\n <id>http://arxiv.org/abs/2602.23492v1</id>\n <title>Automated Extraction of Unstructured Post-SBRT Toxicity Data from Radiology Reports Using Large Language Models</title>\n <updated>2026-02-26T20:54:29Z</updated>\n <link href='https://arxiv.org/abs/2602.23492v1' rel='alternate' type='text/html'/>\n <link href='https://arxiv.org/pdf/2602.23492v1' rel='related' title='pdf' type='application/pdf'/>\n <summary>We evaluated the viability of using a Large Language Model (LLM) to extract patient-specific specific toxicity and progression outcomes from unstructured radiology reports. We retrospectively extracted 160 follow-up CT and PET/CT electronic medical record notes for patients treated with lung stereotactic body radiotherapy (SBRT) at our institution from January 2017 through December 2023. Using the Llama 3.3-70-B-Instruct LLM, we engineered prompts to extract four clinical endpoints from each radiology report: locoregional progression, distant progression, radiation-induced fibrosis, and radiation-induced rib fractures. Progression endpoints were classified as yes, no, or maybe, while fibrosis and rib fractures were binary (yes or no). Ground truth labels were defined using two-grader consensus for the 60-note training set, used for prompt development, and a three-grader majority vote for the 100-note test set. LLM performance was evaluated using sensitivity, specificity, and accuracy. As detailed by our evaluation metrics, the strong performance of our methods demonstrates the viability of using prompt-engineered LLMs to extract radiation-toxicities and progression classification from radiology reports.</summary>\n <category scheme='http://arxiv.org/schemas/atom' term='physics.med-ph'/>\n <published>2026-02-26T20:54:29Z</published>\n <arxiv:primary_category term='physics.med-ph'/>\n <author>\n <name>Justin Pijanowski</name>\n </author>\n <author>\n <name>Yakout Mezgueldi</name>\n </author>\n <author>\n <name>Alan Lee</name>\n </author>\n <author>\n <name>Drew Moghanaki</name>\n </author>\n <author>\n <name>Ricky R. Savjani</name>\n </author>\n <author>\n <name>James Lamb</name>\n </author>\n </entry>"
}