NIDB is a designed to maintain an association between a subject and all of their data, including analyses, regardless of project, which encourages sharing and aggregation of large datasets.
NIDB has the following features, some of which are distinct from other database systems:
- Ease of installation – Within 1 hour, after installing the OS, you can have a working NIDB installation and begin receiving DICOM images and importing data.
- Data stored at your site – You have complete control over your data. The database is installed on your servers and is managed by you. No other site has access to your data and you chose when to share.
- Granular data access – The finest grain access is at the series level, so you can easily run a query such as “find all T1 MRIs from males run under project X, within a date range”
- Simple data access – Data from any query can be downloaded to a local NFS share, local FTP, or remote FTP.
- Simple hierarchy – Subjects → imaging studies → series. Subjects can have multiple imaging studies across different projects, and each study has multiple series. This model allows grouping of data across projects.
- Automated QA – Movement and SNR are automatically calculated for MR series. QA for other modalities can easily be added
- Store any type of data – Absolutely any type of binary data can be stored in the system. Along with the standard MR, CT, PET, EEG, NIDB can store audio, video, genetics files, scanned documents, prior analyses, surgical planning data, and more.
- Automated pipeline analysis – Analysis pipelines can be created to run on any data stored in NIDB. Analysis results (values, images, etc) can be imported back into NIDB and are available as part of the search results.
- Unique ID – Subjects are assigned a unique ID when entered or imported into the system. No matter what project or how many imaging studies they perform, they have the same UID. A standard ‘license plate’ style UID allows 175,760,000 subjects to be entered into the system, with unlimited numbers of imaging studies per subject
- Export to spreadsheet – Most search results, including QA and analysis results can be exported to a sensibly formatted .csv file.
- Groups – Groups of subjects, studies, or series can be created. These groups can be searched and analyzed
- DICOM receiver – Connects directly to an MR, CT, PET, or other DICOM sender to receive and process imaging data. Data are available for searching within 1-2 minutes of a series completion. QA, depending on computational complexity, is available within 2-3 minutes.