1. Structural parameters are superior to eigenvector centrality in detecting progressive supranuclear palsy with machine learning & multimodal MRI
- Author
-
Franziska Albrecht, Karsten Mueller, Tommaso Ballarini, Klaus Fassbender, Jens Wiltfang, Markus Otto, Robert Jech, Mattias L. Schroeter, Adrian Danek, Janine Diehl-Schmid, Holger Jahn, Jan Kassubek, Johannes Kornhuber, Bernhard Landwehrmeyer, Martin Lauer, Johannes Prudlo, Anja Schneider, Albert C. Ludolph, Klaus Fliesbach, Sarah Anderl-Straub, Katharina Brüggen, Marie Fischer, Hans Förstl, Anke Hammer, György Homola, Walter Just, Johannes Levin, Nicolai Marroquin, Anke Marschhauser, Danielé Pino, Magdalena Nagl, Timo Oberstein, Lea Hüper, Maryna Polyakova, Hannah Pellkofer, Tanja Richter-Schmidinger, Carola Rossmeier, Marianna Kulko, Elisa Semler, Annika Spottke, Petra Steinacker, Angelika Thöne-Otto, Ingo Uttner, and Heike Zech
- Subjects
Progressive supranuclear palsy ,Magnetic resonance imaging ,Resting-state functional connectivity ,Voxel-based morphometry ,Eigenvector centrality ,Support vector machine ,Science (General) ,Q1-390 ,Social sciences (General) ,H1-99 - Abstract
Progressive supranuclear palsy (PSP) is an atypical Parkinsonian syndrome characterized initially by falls and eye movement impairment. This multimodal imaging study aimed at eliciting structural and functional disease-specific brain alterations. T1-weighted and resting-state functional MRI were applied in multi-centric cohorts of PSP and matched healthy controls. Midbrain, cerebellum, and cerebellar peduncles showed severely low gray/white matter volume, whereas thinner cortical gray matter was observed in cingulate cortex, medial and temporal gyri, and insula. Eigenvector centrality analyses revealed regionally specific alterations. Multivariate pattern recognition classified patients correctly based on gray and white matter segmentations with up to 98 % accuracy. Highest accuracies were obtained when restricting feature selection to the midbrain. Eigenvector centrality indices yielded an accuracy around 70 % in this comparison; however, this result did not reach significance. In sum, the study reveals multimodal, widespread brain changes in addition to the well-known midbrain atrophy in PSP. Alterations in brain structure seem to be superior to eigenvector centrality parameters, in particular for prediction with machine learning approaches.
- Published
- 2024
- Full Text
- View/download PDF