Inflammatory arthritis comprises a group of diseases in which the immune system attacks the body’s own tissues. Two prevalent types of inflammatory arthritis are rheumatoid arthritis (RA) and... Show moreInflammatory arthritis comprises a group of diseases in which the immune system attacks the body’s own tissues. Two prevalent types of inflammatory arthritis are rheumatoid arthritis (RA) and spondyloarthritis (SpA). Clinical research points to the importance of early diagnosis, as treatment in early disease stages increases chances of better outcome and improved quality of life for patients. To this end, the diagnostic potential of imaging modalities sensitive to local inflammation, such as magnetic resonance imaging (MRI), is of great interest. The goal of this thesis was to develop computer-aided methods for assessment of MRI-detected inflammation with the aim of aiding early diagnosis of inflammatory arthritis. In particular, we focused on the tasks of comparative visualization, automatic quantification, and feature selection. The presented studies showcase the potential of comparative visualization and automatic quantification to overcome the limitations of visual scoring and lay out a fertile ground for future improvements. Additionally, the understanding of the diagnostic role of individual inflammatory features in prediction of RA development is further advanced. Collectively, these findings can help facilitate the use of MRI for early diagnosis of inflammatory arthritis and potentially increase chances of better outcome and quality of life for patients. Show less
Aizenberg, E.; Brinck, R.M. ten; Reijnierse, M.; Helm-van Mil, A.H.M. van der; Stoel, B.C. 2019