Skip to content
Main menu hidden.

Early detection of brain tumors with AI

Research project Each year, about 22,000 people in the EU are diagnosed with a glioma, a malignant type of brain tumor. After diagnosis the median survival time is only 15 months. In a new project we are combining the expertise of brain tumors and AI modelling to find more effective methods of early glioma detection

Large amounts of data about this disease have already been collected by our research group at Umeå University, where we are focusing on understanding the origin of gliomas. We are collecting comprehensive data sets including health data, genomics, metabolites and proteins. We have found genetic variants and metabolites that are associated with increased risk of glioma.

Head of project

Beatrice Melin
Professor, senior consultant (attending) physician

Project overview

Project period:

Start date: 2024-01-01

Participating departments and units at Umeå University

Department of Radiation Sciences

Research area


Project description

Traditional statistical analyses have only marginally improved the early detection of gliomas. This is largely due to having only a small number of
datapoints with predictive value. Detecting gliomas can thus be compared with identifying the proverbial “needle in the haystack”.

Heights AI and PSE Data Security are two companies (from the Netherlands and Switzerland respectively) with expertise in artificial intelligence models for identifying anomalies comparable with a “needle in the haystack”. This is why we’ve decided to join forces and combine our domain knowledge with their modelling skills in a partnership.

Our ultimate goal is to improve the early detection of gliomas and to find better targets for treatment.



Latest update: 2024-01-04