{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Lopez I"],"funding":["NCATS NIH HHS","U.S. Department of Health & Human Services | NIH | National Institute of Allergy and Infectious Diseases (NIAID)"],"pagination":["45"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC11743751"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["8(1)"],"pubmed_abstract":["Large language models (LLMs) with retrieval-augmented generation (RAG) have improved information extraction over previous methods, yet their reliance on embeddings often leads to inefficient retrieval. We introduce CLinical Entity Augmented Retrieval (CLEAR), a RAG pipeline that retrieves information using entities. We compared CLEAR to embedding RAG and full-note approaches for extracting 18 variables using six LLMs across 20,000 clinical notes. Average F1 scores were 0.90, 0.86, and 0.79; inference times were 4.95, 17.41, and 20.08 s per note; average model queries were 1.68, 4.94, and 4.18 per note; and average input tokens were 1.1k, 3.8k, and 6.1k per note for CLEAR, embedding RAG, and full-note approaches, respectively. In conclusion, CLEAR utilizes clinical entities for information retrieval and achieves >70% reduction in token usage and inference time with improved performance compared to modern methods."],"journal":["NPJ digital medicine"],"pubmed_title":["Clinical entity augmented retrieval for clinical information extraction."],"pmcid":["PMC11743751"],"funding_grant_id":["UM1 TR004921","UL1 TR003142","1R01AI17812101"],"pubmed_authors":["Ma SP","Vedula K","Maddali M","Gallo RJ","Narayanan S","Tate S","Lopez I","Swaminathan A","Nateghi Haredasht F","Chen JH","Shah NH","Liang AS"],"additional_accession":[]},"is_claimable":false,"name":"Clinical entity augmented retrieval for clinical information extraction.","description":"Large language models (LLMs) with retrieval-augmented generation (RAG) have improved information extraction over previous methods, yet their reliance on embeddings often leads to inefficient retrieval. We introduce CLinical Entity Augmented Retrieval (CLEAR), a RAG pipeline that retrieves information using entities. We compared CLEAR to embedding RAG and full-note approaches for extracting 18 variables using six LLMs across 20,000 clinical notes. Average F1 scores were 0.90, 0.86, and 0.79; inference times were 4.95, 17.41, and 20.08 s per note; average model queries were 1.68, 4.94, and 4.18 per note; and average input tokens were 1.1k, 3.8k, and 6.1k per note for CLEAR, embedding RAG, and full-note approaches, respectively. In conclusion, CLEAR utilizes clinical entities for information retrieval and achieves >70% reduction in token usage and inference time with improved performance compared to modern methods.","dates":{"release":"2025-01-01T00:00:00Z","publication":"2025 Jan","modification":"2025-04-22T10:32:29.612Z","creation":"2025-04-22T10:32:29.612Z"},"accession":"S-EPMC11743751","cross_references":{"pubmed":["39828800"],"doi":["10.1038/s41746-024-01377-1"]}}