Swedish Medical Language Data Lab
| Reference number | |
| Coordinator | Sahlgrenska Science Park AB |
| Funding from Vinnova | SEK 1 889 349 |
| Project duration | December 2019 - December 2021 |
| Status | Completed |
| Venture | Data-driven innovation |
| Call | Data lab and data factory as national resource 2020 |
Purpose and goal
Focus for Swedish Medical Language Data Lab changed from making medical data available to establishing methods for safe and secure local access to sensitive data. This in turn enabled training of AI models on real patient records with promising results and lead to learnings around the legal considerations for data sharing.
Expected results and effects
The project has been successful and results and learnings are now implemented in regular operations at Region Halland, Sahlgrenska Universityhospital and Public Dental Care. The project partners continue to collaborate on training AI models. There are plans for ethical review, further testing and publication of results for prediction of fall injuries in inpatient care with the ambition to implement it as a decision support tool. The Swedish Health and Care Inspectorate and the Swedish Research Council have shown great interest in lessons learned and results from the project.
Planned approach and implementation
Important success factors have been a close dialogue in the project team with regular meetings and many informal contacts, but above all a great respect for each other´s know-how between the healthcare owners and the domain experts. Something that has been incredibly important is also that the need owners after the first Covid wave have been able to invest considerable resources and in parallel have been driving internal change programmes in their respective organizations with increased data readiness and an information-driven approach as overall objective.