We are funding a project to develop an AI-solution capable of reading Nordic patient records

In addition to the funding, there will be a co-financing requirement for the winning consortia of at least +50%. 

GO TO CALL

The goal of the project is to develop and demonstrate an ethical algorithm capable of processing ethical AI-algorithm to read text and images, both digital and analogue patient records, across medical health record systems and across Nordic borders and Nordic languages. Furthermore, the project should adhere to the protection of privacy of individuals and follow FAIR principles and existing AI guidelines, standards as well as forthcoming EU regulations in the area of AI and healthcare.

“With this project, we want to show what can be done in the area of ethical AI applied to responsible use of health data within the current legal framework”, says Olivia Rekman, Innovation Advisor at Nordic Innovation.

“Furthermore, we also expect that this project will foster Nordic competitive advantage in the area of ethical AI applied to responsible use of health data and that it will inspire future developments in this area”, says Rasmus Malmborg., Senior Advisor at Nordic Innovation.

Background

While registry data is often used in research, this is not as common for patient records. Patient records are often only semi-structured and contain written notes. However, in comparison with registry data, patient records potentially give a broader and deeper context than registry data. When mining these large data sets, it would be useful to have an AI solution.

Ethically applied AI/machine learning-methods to patient records have the potential to radically improve healthcare outcomes. Both by improving workflows and by efficiently detecting clinically relevant information hidden in the massive amount of data in patient records, AI holds great potential in improving life expectancy and quality of life for the individual patient. Analyzing this type of medical data at scale could result in novel methods of identifying and treating diseases and illnesses.

There are several barriers (such as technical, ethical and legal) to unlocking the great potential that Nordic health data holds in providing data insights from context-rich patient records. This call is targeted at tackling the technical challenges within the current legal framework and changes to the legal framework that occurs during the full project period.

One of the main technical challenges is the secure access and exchange of data.

For an AI solution applied to patient records to be able to maximize the output and benefit of available data, it needs to be able to function across systems and languages, while adhering to the highest of standards and guidelines concerning ethical and privacy-related aspects.

For the healthcare system, having AI solutions that can interpret and model based on patient records across languages and systems, improves the ability to utilize the information in patient records. The larger pools and higher quality of data strengthen the usability of the developed solutions.

A cross-sectoral collaboration

The call is one out of two calls in a cross-sectoral collaboration between the Life Science & Health Tech and AI & Data programs in addition to the Nordic Council of Ministers for Digitalization, with the aim of innovating the technical AI solution The aim of this call is to demonstrate the added value from utilizing patient summaries in innovating new solutions for solving health care challenges. Together, the two resulting projects aim at propelling the Nordic region as leading in ethical AI and responsible use of health data, by demonstrating the potential for value creation for patients, healthcare systems and Nordic businesses.