The MIMIC-CXR Database

MIMIC-CXR is a large, publicly-available database comprising of de-identified chest radiographs from patients admitted to the Beth Israel Deaconess Medical Center between 2011 and 2016. The dataset contains 371,920 chest x-rays associated with 227,943 imaging studies. Each imaging study can pertain to one or more images, but most often are associated with two images: a frontal view and a lateral view. Images are provided with free-text radiology reports. All images and reports have been de-identified to protect patient privacy. The dataset is made freely available to facilitate and encourage a wide range of research in medical computer vision. All data is organized into a hierarchical directory structure. At the highest level is a set of 10 folders (p10 - p19), each with ~6,500 sub-folders. Sub-folders are named according to the patient identifier, and contain free-text reports and DICOM files for all studies for that patient. All imaging studies from an individual patient are stored in the same folder under an anonymized patient identifier prefixed with the letter 'p'. All imaging studies are prefixed with the letter 's' followed by the anonymous study identifier. DICOM filenames are 40 character hashes with a dash every 8 characters.

The MIMIC-CXR Database, although de-identified, still contains detailed information regarding the clinical care of patients, and must be treated with appropriate care and respect. Researchers seeking to use the full Clinical Database must formally request access. Please login to the new PhysioNet platform at the PhysioNet login page. Then, see the credentialing instructions for more information on the credentialing process.

The latest version of the database can be accessed at: https://physionet.org/content/mimic-cxr/

Detailed documentation on the dataset is available online: https://mimic-cxr.mit.edu/