Authors: Mohamed Sobhi Jabal, Pranav Warman, Jikai Zhang, Kartikeye Gupta, Ayush Jain, Maciej Mazurowski, Walter Wiggins, Kirti Magudia, Evan Calabrese
Published: 2025-03-12
DOI: 10.1148/ryai.240551
Source: Full article
The best-performing open-weight language models had high accuracy for automated extraction of structured clinical data points from unstructured radiology and pathology reports, with retrieval-augmented generation improving performance for complex reports.