Due to copyright issues, an electronic copy of the thesis must be ordered from the faculty. For the faculty to have time to process the order, the order must be received by the faculty at the latest 2 days before the public defence. Orders received later than 2 days before the defence will not be processed. After the public defence, please address any inquiries regarding the thesis to the candidate.
Trial Lecture – time and place
See Trial Lecture.
Adjudication committee
- First opponent: University Lecturer Hanna Jokinen, University of Helsinki
- Second opponent: Associate Professor Olga Therese Ousdal, University of Bergen
- Third member and chair of the evaluation committee: Professor Mathias Toft, University of Oslo
Chair of the Defence
Professor II Espen Dietrichs, University of Oslo
Principal Supervisor
Professor Mona K. Beyer, University of Oslo
Summary
Can brain imaging be used to distinguish stroke survivors who are left with cognitive problems years after the stroke from those who run free?
The aim of this thesis was to identify MRI markers for the prediction of post stroke neurocognitive disorder (NCD). By following stroke patients over 3 years and through the use of methods in machine learning, we set out to investigate the contribution of vascular vs. neurodegenerative processes; the potential difference between the two hemispheres; and lastly the use of a brain age marker.
We found that rapid (<3 months) onset of NCD could be predicted with 87.6% accuracy if combining vascular-, neurodegenerative-, and stroke characteristic factors. The top factors, by weighting, were stroke lesion volume, white matter hyperintensity volume, and left occipital cortical thickness. In contrast to previous literature we found that vascular factors were more important than neurodegenerative ones, indicating the disorder is more complex than earlier suggested.
Next, the ipsilesional hemisphere atrophied more than the contralesional. Interestingly, the effect was more prominent in the right hemisphere, indicating that perhaps the left hemisphere is more resilient to stroke. We also found that a greater degree of atrophy was associated with worse cognition over time, especially if in the medial temporal lobe or with larger stroke volumes.
Lastly, a high brain age gap (BAG) indicates an older appearing brain compared to chronological age. We found that a high BAG was associated with NCD over time. A younger appearing brain seems to reflect a resilience towards cognitive decline, demonstrating that BAG can be used as a predictive marker for post stroke NCD.
Post stroke NCD is complex and the outcome seems in part dependent on brain health. This points to the importance of taking proactive measures. Using neuroimaging and developing robust prediction models is of huge advantage in this work and should be in focus in the years to come.
Additional information
Contact the research support staff.