AI-based tools for automated analysis of PET/CT images
| Reference number | |
| Coordinator | Region Skåne - VO Bild och Funktion |
| Funding from Vinnova | SEK 1 000 000 |
| Project duration | July 2020 - December 2021 |
| Status | Completed |
| Venture | Medtech4Health innovators |
| Call | Medtech4Health: Innovators in Healthcare and Care 2020 |
Important results from the project
** Denna text är maskinöversatt ** The aim of the project is to develop tools based on artificial intelligence for the development of image biomarkers from a diagnostic image method, positron emission tomography combined with computed tomography (PET-CT), which has been validated and verified by several hospitals and that we started a discussion about commercialization. PET-CT is used to evaluate patients with cancer, and can be crucial in the choice of treatment. All four work packages have been delivered.
Expected long term effects
** Denna text är maskinöversatt ** The project application describes four work packages. The work has resulted in 1) a report describing user eligibility requirements on an AI tool 2) a prototype for automatic analysis of PET-CT-bilder in lymphoma and lung cancer 3) a report describing the validation and verification process and 4) preparation for contact with potential partners. We have made the AI prototype for lung cancer freely available to other researchers. This facilitates validation of the product through collaboration with other hospitals.
Approach and implementation
** Denna text är maskinöversatt ** This is a collaborative project between Skåne University Hospital, Sahlgrenska University Hospital and Eigenvision AB. The steering group has had weekly meetings as well as regular meetings with doctors at collaborating hospitals. The combination of medical expertise in the field of AI analysis by PET-CT-bilder from two different hospitals and technical expertise in the field of development of AI tools for medical image analysis has been crucial for a successful design and execution.