Quantitative Imaging Biomarkers for Hippocampus-based Dementia Prediction
Institution: | Erasmus University Rotterdam |
---|---|
Department: | |
Year: | 2017 |
Posted: | 02/01/2018 |
Record ID: | 2154902 |
Full text PDF: | http://hdl.handle.net/1765/102960 |
markdownabstractThis thesis is about hippocampus-based quantitative imaging biomarkers for the prediction of dementia. The research focussed on the following aspects of the biomarkers: 1) robust segmentation of brain structures on scans with varying acquisition protocols, 2) the predictive value of hippocampal volume, shape, and texture for dementia, 3) spatially aware regression of shape data using P-spline based regression, and 4) a workflow framework for large-scale extraction of biomarkers.