Information driven healthcare
In the future, healthcare will look quite different from how it does today. Demographic change will increase the demand for care and a larger proportion of people will suffer from chronic conditions. It is clear that the transformation needed in the healthcare sector cannot be achieved simply by providing more resources. New technical solutions and forms of cooperation and different working methods are required. Artificial intelligence has the potential to play a key role in this transformation for systems and for individuals.
In order to enable AI to be introduced into healthcare on a national level, more knowledge is needed about the data, the legal issues and the implementation. It would not be sustainable for all the regions to develop solutions individually. Greater coordination and collaboration are required. In addition, the solutions that have been created on a local or regional level must be distributed and implemented throughout the country. More cooperation between different regions and caregivers will lead to better use of healthcare resources. No public bodies should need to reinvent the wheel.
One area of the project involves identifying and consolidating important examples, while also sharing experience and knowledge with interested parties that want to follow and learn from the project. The core project team consists of AI Sweden, the Swedish Association of Local Authorities and Regions (SKR), Region Halland, Halmstad University and Karolinska University Hospital.
Region Halland and Halmstad University have focused on structuring the large volumes of data often found in healthcare systems. This has been a key step towards being able to work with the data and produce algorithms for predicting mortality, among other things. As part of the project, Region Halland and Halmstad University will also share experiences, develop solutions for federated learning and carry out research into the implementation of information-driven care. In simple terms, federated learning involves training algorithms using data in different decentralised locations without sharing data samples.
Unlike many AI projects in healthcare, this project focuses not only on using new technologies but also on creating solutions that can be implemented on a national level. Greater cooperation on decisive issues relating to the development and implementation of AI solutions will lead to a more efficient use of healthcare resources. In the long term, the solutions have the potential to allow for more personalised care and the creation of a data-driven healthcare system. The knowledge and the solutions that are developed will be able to support the move towards more individualised healthcare and to change the perspective from the past (how was it?) to the future (how will it be?).
A part of the national Vision-Driven Healthcare programme
The project involves the long-term establishment of an innovation environment. The work is being funded for five years by Vinnova, Sweden’s innovation agency, and the project partners in the national Vision-Driven Healthcare programme in Sweden. The environment is one of five innovation environments for health that have been given priority in Sweden. A large number of new projects and preliminary studies will be launched as part of the innovation environment. One result of the project is a new cooperation between Region Halland and Karolinska University Hospital which involves an algorithm for predicting mortality that has been developed in Halland being tested and evaluated at the hospital.
Cooperating in the initial phase are AI Sweden, SKR, Region Halland, Halmstad University, and Innovationsplatsen, Karolinska University Hospital. The project will last for five years and active work to invite and identify new participants takes place continuously.