Your browser doesn't support javascript. This means that the content or functionality of our website will be limited or unavailable. If you need more information about Vinnova, please contact us.

Swedish Medical Language Data Lab

Reference number
Coordinator Sahlgrenska Science Park AB
Funding from Vinnova SEK 1 913 272
Project duration December 2019 - December 2021
Status Ongoing
Venture Data-driven innovation
Call Data lab and data factory as national resource 2020

Purpose and goal

Svenskt Medicinskt Språkdatalabb is to be a national experimental workshop with the aim to make medical language data sets available for use in health care, dental care and data-driven research and innovation in the life science sector. The project has a clear user focus and goal to develop language models as well as technical, legal and ethical processes that enable the sharing of sensitive training data for AI solutions in the life science sector.

Expected results and effects

The project is to create: A national Swedish medical language database with a long-term plan for ownership and operation. A growing knowledge node where need owners, experts and stakeholders meet and share knowledge. Processes and licensing models for access to data sets and models that enable application of medical natural language processing for health care and life science industry.

Planned approach and implementation

The project is to be implemented as follows: In-depth requirements and needs analysis. Inventory and evaluation of existing data and models from a needs perspective. Technical adaptation of data. Modeling of models. Develop trained models for language understanding of Swedish medical text including documented evaluations of method and model choices. Sharing and making data available. Models will be published as open source code in all instances possible. Distribution of results.

The project description has been provided by the project members themselves and the text has not been looked at by our editors.

Last updated 7 May 2021

Reference number 2019-05156

Page statistics