This site provides information on datasets shared on the AIDA data hub through search engine optimized DOI landing pages, and also in machine-readable form through an API.
Please report any errors on this website to our GitHub Pages issue tracker, or submit a pull request.
For questions regarding data please contact the AIDA data director.
For all other inquiries please contact AIDA.
What are DOIs and dataset registers?
The AIDA data hub facilitates sharing datasets for machine learning in medical image diagnostics. AIDA uses Digital Object Identifiers (DOI) to increase exposure and impact of the shared datasets. These DOIs help make the datasets more findable, accessible, interoperable, and reusable (FAIR).
Example: 10.23698/aida/ctpa → https://doi.org/10.23698/aida/ctpa
Thanks to the global DOI system, following a link to a DOI will always take you the official landing page with the latest updated official information on that dataset.
AIDA has its own DOI prefix provided by DataCite through the Swedish National Data Service (SND). The AIDA data hub and its datasets are discoverable through the global re3data.org registry of research data repositories:
You can find us and other data sources relevant to your research if you browse by subject, for example:
- Radiology and Nuclear Medicine
- Pathology and Forensic Medicine
- Artificial Intelligence, Image and Language Processing.
Access and privacy
AIDA is a collaboration arena for academia, industry and healthcare and supports OpenScience and FAIR data. AIDA can facilitate large scale data exports for research from clinical production systems, and can host research data for sharing. AIDA uses data mainly from medical imaging, and therefore has an obligation to adequately safeguard the privacy of the individuals concerned. Therefore AIDA only shares data that is ethically approved for sharing, and when a contractual agreement is in place that includes non-disclosure of data, such AIDA partner contracts for innovation projects, clinical- or technical fellowships, on-site development projects, or network partnerships. AIDA also facilitates contacts with data controllers for datasets shared on AIDA for data sharing options outside of AIDA.
Please see the AIDA GDPR policy for more details. AIDA is planning a revision of its platform and will update this section as extended capabilities become available.
Anonymous vs identifiable data
Currently, AIDA exclusively uses anonymized data. However to facilitate export and work with larger amounts of data, AIDA is planning to establish a platform that is secure enough to store, share and process also identifiable data.
This site uses the following free and open services and technologies:
- DataCite - DOI minting and management for research datasets.
- SND - Swedish member of DataCite.
- re3data.org - Dataset register discoverability.
- GitHub Pages - Hosting, issue tracking and collaborative development.
- Jekyll - Static website generator.
- ViewerJS - Interactive image viewer.
- AnchorJS - Linkable headings.
- DataTables - Sortable and searchable tables.