| Quick Links |  |
|
|---|---|---|
| Getting Access | ||
| Waveform Database Overview | ||
| Clinical Database Overview | ||
| Waveform Data | ||
| Clinical Data | ||
| MIMIC II Statistics | ||
| MIMIC II FAQs |
The MIMIC II (Multiparameter Intelligent Monitoring in Intensive Care) Database contains comprehensive clinical data from tens of thousands of Intensive Care Unit (ICU) patients*. Data were collected between 2001 and 2008 from a variety of ICUs (medical, surgical, coronary care, and neonatal) in a single tertiary teaching hospital. The database contains clinical data from bedside workstations as well as hospital archives. The database also includes thousands of records of continuous high-resolution physiologic waveforms and minute-by-minute numeric time series (trends) of physiologic measurements. A significant fraction of the waveform records are matched to corresponding clinical data (for more information please see Record Matching). The database is thoroughly de-identified (all PHI has been removed and all dates have been changed).
The database is distributed in two sections: the Clinical Database, and the Waveform Database. Access to the MIMIC II Database is available free of charge to qualified researchers who obtain human subjects training and sign a simple data use agreement (see Getting Access).
Resources
- Introductory Material
- Web exploration tools (resources hosted by the MIMIC II project)
- Query Builder (Clinical Data, login required).
- User Guide and Documentation for the MIMIC II Database (html) (pdf)
- MIMIC II on PhysioNet
- Waveform Data (binary files in PhysioBank formats; no registration required).
- Clinical Data (tab-delimited text files; restricted access requiring a key to decrypt).
- PhysioBank ATM (view and manipulate waveform data in a web browser; no registration required).
- Using the MIMIC II Database (tutorial on using the database with PhysioNet tools).
- PhysioNet/CinC Challenge 2009: Predicting Acute Hypotensive Episodes (using MIMIC II data)
*The MIMIC II Database is the product of an ongoing NIH-sponsored Bioengineering Research Partnership (NIBIB BRP R01EB001659), that combines the resources of a interdisciplinary team from academia (MIT), industry (Philips Healthcare), and clinical medicine (Beth Israel Deaconess Medical Center) to develop and evaluate advanced ICU patient monitoring and decision support systems.