MIMIC-IV-Note: Deidentified free-text clinical notes
Open accessThe DOI matches its DataCite record; none of the 8 sources its claims cite is checked for support yet.
Abstract
The advent of large, open access text databases has driven advances in state- of-the-art model performance in natural language processing (NLP). The relatively limited amount of clinical data available for NLP has been cited as a significant barrier to the field's progress. Here we describe MIMIC-IV-Note: a collection of deidentified free-text clinical notes for patients included in the MIMIC-IV clinical database. MIMIC-IV-Note contains 331,794 deidentified discharge summaries from 145,915 patients admitted to the hospital and emergency department at the Beth Israel Deaconess Medical Center in Boston, MA, USA. The database also contains 2,321,355 deidentified radiology reports for 237,427 patients. All notes have had protected health information removed in accordance with the Health Insurance Portability and Accountability Act (HIPAA) Safe Harbor provision. All notes are linkable to MIMIC-IV providing important context to the clinical data therein. The database is intended to stimulate research in clinical natural language processing and associated areas.
Overview
CoChat Summary
## TL;DR
MIMIC-IV-Note is a deidentified free-text clinical notes dataset linked to MIMIC-IV, intended to support clinical NLP research. It contains 331,794 discharge summaries from 145,915 patients and 2,321,355 radiology reports from 237,427 patients, with PHI removed under HIPAA Safe Harbor.
## Key findings
- The database includes 331,794 deidentified discharge summaries from 145,915 patients admitted to the hospital and emergency department at Beth Israel Deaconess Medical Center.
- It also includes 2,321,355 deidentified radiology reports from 237,427 patients.
- All notes were deidentified by a custom rule-based approach combined with a neural network trained for deidentification; annotations were unioned and PHI was removed.
- Previous work reported the deidentification approach had 99.9% sensitivity for radiology reports, and a manual review of discharge summaries found no PHI.
## Methods
- The resource includes notes for patients in MIMIC-IV; all MIMIC-IV inclusion criteria also apply to MIMIC-IV-Note, and notes are generally included only if they occur within one year of a patient encounter.
- Free-text notes were acquired from the hospital system and deidentified using a rule-based method plus a neural network; PHI was replaced with exactly three underscores.
- The dataset is organized into four tables: discharge, discharge_detail, radiology, and radiology_detail, with notes distributed as CSV files and each row corresponding to a unique note.
## Limitations
- The paper states that release v2.2 changed only project metadata, so it does not report a new substantive evaluation of the dataset in that release.
- The text makes evident that this is a restricted-access resource, available only to credentialed users who complete required training and sign the DUA (evident from the access section).
- Not stated in the paper.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 Valid · 2 Not valid · 6 Not checked · 25 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The database also contains 2,321,355 deidentified radiology reports for 237,427 patients.”
plainn = 237,427- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In general, only notes occurring within one year of a patient encounter are included in the database, where an encounter is defined as an emergency department or hospital stay.”
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“MIMIC-IV-Note contains 331,794 deidentified discharge summaries from 145,915 patients admitted to the hospital and emergency department at the Beth Israel Deaconess Medical Center in Boston, MA, USA.”
plainn = 145,915- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“Annotations from the two approaches were unioned, and identified instances of protected health information (PHI) were removed from the notes.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“Free-text notes were acquired from the hospital system and deidentified using a custom rule-based approach combined with a neural network trained for deidentification.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
6 more claims
Create a free account to see all 33 claims and every source verdict.
- Probably not a substantial claim. 20
- It may state two separate claims as one. 5
Citations
Cited by
Gaber, F., Shaik, M., Allega, F., Bilecz, A. J., Busch, F., Goon, K., Franke, V., & Akalin, A. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-01684-1 · Cited by 165
An agentic system for rare disease diagnosis with traceable reasoning
Zhao, W., Wu, C., Fan, Y., Qiu, P., Zhang, X., Sun, Y., Zhou, X., Zhang, S., Peng, Y., Wang, Y., et al. (2026) · Nature · doi:10.1038/s41586-025-10097-9 · Cited by 62
A Unified Review of Deep Learning for Automated Medical Coding
Ji, S., Li, X., Sun, W., Dong, H., Taalas, A., Zhang, Y., Wu, H., Pitkänen, E., & Marttinen, P. (2024) · ACM Computing Surveys · doi:10.1145/3664615 · Cited by 82
Multimodal risk prediction with physiological signals, medical images and clinical notes
Wang, Y., Yin, C., & Zhang, P. (2024) · Heliyon · doi:10.1016/j.heliyon.2024.e26772 · Cited by 66
MIMIC-IV-ECG: Diagnostic Electrocardiogram Matched Subset
Gow, B., Pollard, T., Nathanson, L. A., Johnson, A., Moody, B., Fernandes, C., Greenbaum, N., Waks, J. W., Eslami, P., Carbonati, T., et al. (2023) · PhysioNet · doi:10.13026/4nqg-sb35 · Cited by 54
Self-citation
Metadata
- Authors
- Johnson, A., Pollard, T., Horng, S., Celi, L. A., & Mark, R.
- Venue
- PhysioNet
- Publisher
- PhysioNet
- Published
- 2023
- OpenAlex
- W6941414102
- Cited by
- 111
- References
- 0
- Citations by year
- 2026: 75 · 2025: 25 · 2024: 10 · 2023: 1
- Type
- dataset
- Language
- en
- Open access
- green
- doi.org
- Locations
- PhysioNet
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Health Sciences › Medicine › Health Informatics › Biomedical Text Mining and Ontologies
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Keywords
- discharge summaries, radiology reports, clinical natural language processing, MIMIC-IV
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-06, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-06 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: MIMIC-IV-Note: Deidentified free-text clinical notes. https://app.cochat.ai/papers/10.13026/1n74-ne17
- The paper
- Johnson, A., Pollard, T., Horng, S., Celi, L. A., & Mark, R. (2023). MIMIC-IV-Note: Deidentified free-text clinical notes. PhysioNet. https://doi.org/10.13026/1n74-ne17
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.