Machine learning analysis of cerebral microstructure to find biomarkers of vertigo

Details

Supervisors

Macq, Benoît ; Dricot, Laurence

Faculty

Ecole polytechnique de Louvain

Degree label

Master [120] : ingénieur civil biomédical, à finalité spécialisée

Abstract

Machine learning and tractography are methods that have gained interest in the previous years to study brain microstructure. The purpose of this master thesis is to use those tools to find biomarkers for vertigo in hearing-impaired patients, while maintaining a maximum interpretability of the obtained results. A workflow was created based on previously existing methods to produce bundles and analyse DTI measures in precise locations of these bundles. Those tract-wise analysis were also compared with regions-wise analysis of the microstructural metrics, structural lateralization of key bundles, and FSL's statistical analysis.