LabQAR (Laboratory Question Answering with Reference Ranges) Data Set

LabQAR (Laboratory Question Answering with Reference Ranges), a manually curated dataset containing multiple-choice questions about 550 lab tests with comprehensive reference ranges sourced from trusted medical resources with annotations on reference ranges, specimen types, and other factors impacting interpretation. We also assess the performance of several large language models (LLMs), including LLaMA 3.1, GatorTronGPT, GPT-3.5, GPT-4, and GPT-4o, in predicting reference ranges and classifying results as normal, low, or high.