MIMIC-IV Demo 2.2 TABULAR patient-level
PhysioNet · ODC-BY-1.0 (PhysioNet Open Data)
De-identified ICU + hospital EHR tables (labs, admissions, chart events, prescriptions). Rich clinical schema with well-known join keys — a strong fit for clinical NLP, phenotyping, and structured-EHR model development.
Potential training uses
- Clinical phenotyping & risk models
- Structured-EHR representation learning
- Lab-result and medication reasoning tasks
- Description
- De-identified demo of MIMIC-IV hospital and ICU clinical tables as gzip-compressed CSV files.
- Source
- https://physionet.org/content/mimic-iv-demo/2.2/
- Tags
- clinicalehrpatient-levelcsv
Request access
Listing details and schema are open for evaluation. Request access to unlock representative previews (sample rows and image thumbnails).
Snapshot & provenance
- Snapshot kind
- TRUSTED_LOCAL read-only reference (not copied)
- State
- PUBLISHED
- Files
- 35
- Package SHA-256
- — (trusted local seed)
- Published
- 2026-07-11T08:58:35.602565+00:00
Quality overview
0 error · 0 warning · 0 info
Assets (35)
CSV_TABLE · 32
| Name | Source path | Files | Rows | Quality |
|---|---|---|---|---|
| patients.csv.gz | hosp/patients.csv.gz | 1 | 100 | OK |
| poe_detail.csv.gz | hosp/poe_detail.csv.gz | 1 | 3795 | OK |
| microbiologyevents.csv.gz | hosp/microbiologyevents.csv.gz | 1 | 2899 | OK |
| omr.csv.gz | hosp/omr.csv.gz | 1 | 2964 | OK |
| procedures_icd.csv.gz | hosp/procedures_icd.csv.gz | 1 | 722 | OK |
| prescriptions.csv.gz | hosp/prescriptions.csv.gz | 1 | 18087 | OK |
| chartevents.csv.gz | icu/chartevents.csv.gz | 1 | 668862 | OK |
| procedureevents.csv.gz | icu/procedureevents.csv.gz | 1 | 1468 | OK |
| demo_subject_id.csv | demo_subject_id.csv | 1 | 100 | OK |
| admissions.csv.gz | hosp/admissions.csv.gz | 1 | 275 | OK |
| d_hcpcs.csv.gz | hosp/d_hcpcs.csv.gz | 1 | 89200 | OK |
| d_icd_diagnoses.csv.gz | hosp/d_icd_diagnoses.csv.gz | 1 | 109775 | OK |
| d_icd_procedures.csv.gz | hosp/d_icd_procedures.csv.gz | 1 | 85257 | OK |
| d_labitems.csv.gz | hosp/d_labitems.csv.gz | 1 | 1622 | OK |
| diagnoses_icd.csv.gz | hosp/diagnoses_icd.csv.gz | 1 | 4506 | OK |
| drgcodes.csv.gz | hosp/drgcodes.csv.gz | 1 | 454 | OK |
| emar.csv.gz | hosp/emar.csv.gz | 1 | 35835 | OK |
| emar_detail.csv.gz | hosp/emar_detail.csv.gz | 1 | 72018 | OK |
| hcpcsevents.csv.gz | hosp/hcpcsevents.csv.gz | 1 | 61 | OK |
| provider.csv.gz | hosp/provider.csv.gz | 1 | 40508 | OK |
| labevents.csv.gz | hosp/labevents.csv.gz | 1 | 107727 | OK |
| pharmacy.csv.gz | hosp/pharmacy.csv.gz | 1 | 15306 | OK |
| poe.csv.gz | hosp/poe.csv.gz | 1 | 45154 | OK |
| services.csv.gz | hosp/services.csv.gz | 1 | 319 | OK |
| transfers.csv.gz | hosp/transfers.csv.gz | 1 | 1190 | OK |
| caregiver.csv.gz | icu/caregiver.csv.gz | 1 | 15468 | OK |
| d_items.csv.gz | icu/d_items.csv.gz | 1 | 4014 | OK |
| datetimeevents.csv.gz | icu/datetimeevents.csv.gz | 1 | 15280 | OK |
| icustays.csv.gz | icu/icustays.csv.gz | 1 | 140 | OK |
| ingredientevents.csv.gz | icu/ingredientevents.csv.gz | 1 | 25728 | OK |
| inputevents.csv.gz | icu/inputevents.csv.gz | 1 | 20404 | OK |
| outputevents.csv.gz | icu/outputevents.csv.gz | 1 | 9362 | OK |
DOCUMENT · 3
| Name | Source path | Files | Rows | Quality |
|---|---|---|---|---|
| LICENSE.txt | LICENSE.txt | 1 | — | OK |
| SHA256SUMS.txt | SHA256SUMS.txt | 1 | — | OK |
| README.txt | README.txt | 1 | — | OK |
Annotations
HUMAN_CURATED
summary
· APPROVED
De-identified patient-level clinical data (hospital + ICU modules). Shared keys such as subject_id / hadm_id / stay_id enable candidate joins.
HUMAN_CURATED
customer_summary
· APPROVED
De-identified ICU + hospital EHR tables (labs, admissions, chart events, prescriptions). Rich clinical schema with well-known join keys — a strong fit for clinical NLP, phenotyping, and structured-EHR model development.
HUMAN_CURATED
training_uses
· APPROVED
Clinical phenotyping & risk models
Structured-EHR representation learning
Lab-result and medication reasoning tasks