15 September 2020

The most important factor for training world-class artificial intelligence (AI) is access to massive amounts of data. High-quality medical data is particularly hard to come by as it often requires time-consuming processing by specialists to be useful. There is also an insecurity in the community about how to share data in a secure, legal and ethical way.

A window from the data hub.
A data set entry in the AIDA Data Hub.Images showing histology tissue from breast cancer axillary lymph nodes. from
The AIDA Data Hub provides a platform where researchers can gather and share large volumes of medical image data. So far, more than 5 TB image data from radiology and pathology has been added to the data hub.

-AI has good potential to help with many of the challenges that exist in healthcare. AI solutions perform better the more different types of data they are trained on and therefore access to data from different parts of the world is essential, says AIDAs Data Director Joel Hedlund.

The data hub has already facilitated sharing of data sets from Swedish researchers to six other researcher groups in four different countries (Spain, China, South Korea and United Kingdom) and more is in the pipeline.

- We want the data that we put a lot of resources into collecting to come to use as much as possible. We ensure this by making it available to other researchers in a legal and ethical way, Joel Hedlund explains.

Currently, AIDA exclusively uses anonymized data. However, to facilitate export and work with larger amounts of data, AIDA is developing a platform that is secure enough to store, share and process also identifiable data.

FAIR Data

AIDA supports Open Science which means that you are transparent about what you have done and which data you have used. One way to work with Open Science is to follow the FAIR data guidelines.

Findable, Accessible, Interoperable and Reusable (FAIR) data means that you make it possible for other researchers to find and use your data in their research.

In 2020, the European Open Science Cloud (EOSC)-Nordic carried out an independent evaluation of FAIR readiness among 98 data repositories in the Nordics (results to be published). The AIDA Data Hub was included in this evaluation and was shown to be among the 12% best scoring repositories in the sample.

Analytic Imaging Diagnostics Arena, AIDA

AIDA is a Swedish arena for research and innovation on artificial intelligence, AI, for medical image analysis. Here, academia, healthcare and industry will meet to translate technical advances in AI technology into patient benefit in the form of clinically useful tools. The arena is national but has its physical base at CMIV at Linköping University. AIDA is an initiative within the Strategic innovation program Medtech4Health, jointly supported by VINNOVA, Formas and the Swedish Energy Agency.

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

Contact

Latest news from LiU

Closeup of small pieces of liver in a petri dish.

A liver biopsy may predict spread of pancreatic cancer

Microscopic changes in the liver can be used to predict spread of pancreatic cancer. The discovery may provide new ways of predicting the course of the disease and preventing pancreatic cancer from spreading to other organs.

Woman with arms crossed.

She wants to make robots behave better

Researcher Hannah Pelikan believes that we will see increased conflicts between humans and robots in the future.  In her research, she films everyday encounters between humans and machines to see what happens.

Two men in a computer server hall.

International collaboration lays the foundation for AI for materials

AI is accelerating the development of new materials. Large-scale use and exchange of data on materials is facilitated by a broad international standard. A major international collaboration now presents an extended version of the OPTIMADE standard.