Please read Dr. Geller’s article in BMC Medical Informatics and Decision Making titled, “Mining of EHR for interface terminology concepts for annotating EHRs of COVID patients.“
Electronic Health Records (EHRs) provide a systematic account of patient health information to authorized users in a real time setting in a secure manner. EHRs include information related to patient demographics, past medical history, medications, vital signs, tests, day-to-day progress, discharge summaries, etc. An EHR stores information in both structured forms (e.g., medication lists, numeric test results, demographic data) and unstructured text (e.g., admission notes, progress notes, discharge summaries). Extensive research has been conducted utilizing the structured part of EHRs, which is easily accessible with diagnostic/procedure/medication terms which correspond to concepts from standard terminologies. Analyzing unstructured textual data in EHRs for extracting concepts related to medical problems, procedures, treatments, etc., is a widely researched topic in clinical natural language processing (NLP). Conventionally, medical text is manually annotated, e.g., to generate data for training different machine learning models. To read the full article.
Mining of EHR for interface terminology concepts for annotating EHRs of COVID patients. Keloth VK, Zhou S, Lindemann L, Zheng L, Elhanan G, Einstein AJ, Geller J, Perl Y. BMC Med Inform Decis Mak. 2023 Feb 24;23(Suppl 1):40. PMID: 36829139 PMCID: PMC9951157 DOI: 10.1186/s12911-023-02136-0