MIMIC-IV-Note: Deidentified free-text clinical notes

The advent of large, open access text databases has driven advances in state-of-the-art model performance in natural language processing (NLP). Therelatively limited amount of clinical data available for NLP has been cited asa significant barrier to the field's progress. Here we describe MIMIC-IV-Note:a collection of deidentified free-text clinical notes for patients included inthe MIMIC-IV clinical database. MIMIC-IV-Note contains 331,794 deidentifieddischarge summaries from 145,915 patients admitted to the hospital andemergency department at the Beth Israel Deaconess Medical Center in Boston,MA, USA. The database also contains 2,321,355 deidentified radiology reportsfor 237,427 patients. All notes have had protected health information removedin accordance with the Health Insurance Portability and Accountability Act(HIPAA) Safe Harbor provision. All notes are linkable to MIMIC-IV providingimportant context to the clinical data therein. The database is intended tostimulate research in clinical natural language processing and associatedareas.