Access to large amounts of data is the most important factor for the successful creation of a world-class artificial intelligence (AI). High-quality medical data is particularly difficult to obtain because the processing required is often time-consuming and requires expertise to be truly useful. There is also uncertainty in the profession about how data can be shared legally and ethically.

AIDA’s Data Hub provides a platform where researchers can collect and share large amounts of medical image data. More than 5 TB of image data from radiology and pathology have so far been added to the data hub.

“AI has great potential to help with many of the challenges that healthcare faces. AI solutions work better the more different types of data they are trained on. Access to data from different parts of the world is therefore crucial,” says Joel Hedlund, Data Director at AIDA.

The data hub has already helped Swedish researchers to share their data with six different research groups in four countries (Spain, the UK, China and South Korea), and further exchanges are underway.

“We want our data, that we’ve put so much of our resources into collecting, to be as useful as possible. We enable this by making it available to other researchers in a legal and ethical manner,” Joel Hedlund explains.

So far, the data in the data hub consists only of anonymised data, but to enable sharing and work with larger data volumes, AIDA is developing a platform that will be secure enough to handle identifiable data as well.

FAIR Data

AIDA supports Open Science, which calls on all researchers to be open about what they have done in their research and what data they have used. One way to work with Open Science is to follow the FAIR data guidelines.

Findable, Accessible, Interoperable and Reusable (FAIR) data entails making it possible for other researchers to find and use our data in their research.

In 2020, the European Open Science Cloud (EOSC) -Nordic conducted an independent evaluation of how good different data sources in the Nordic region are at being FAIR. AIDA Data Hub was one of the 98 data sources included and proved to be among the 12% highest ranked in the sample.

The AIDA Data Set Register: https://datasets.aida.medtech4health.se