Multimodal imaging measures to improve dementia diagnosis

Summary The aim of this study is to study the influence morphological and longitudinal measures to improve dementia diagnosis.

Project Details

Mild cognitive impairment (MCI) is generally regarded as the prodromal stage of Alzheimer’s disease. Magnetic resonance imaging (MRI) is used to extract different indexes from various morphological analyses that reflect intrinsic properties of the cerebral cortex. The best algorithms have yielded an accuracy of 63% and an AUC of 78.8%. Different indexes can be extracted from various morphological analyses and used in machine learning approaches. Further, multimodal imaging measures such as PET, longitudinal imaging measures, and cognitive measures can also be used to improve the dementia diagnosis.

The current study will use imaging and clinical data from a publicly available dataset such as The Australian Imaging Biomarkers and Lifestyle Study of Ageing (AIBL). The aim of this study is to study the influence morphological and longitudinal measures to improve dementia diagnosis.  The student will be responsible for the development of the proposal and refinement of study hypotheses, conducting a literature review, processing of brain imaging scans and performing statistical analyses. Publication of results is expected at the end of the project.