Proposal Description

Neuroimaging is a powerful tool that can help early detection, diagnosis, treatment and prognosis of various brain conditions such as Alzheimer's disease, dementia, and mild traumatic brain injury. There are various forms of neuroimage data such as data produced by Magnetoencephalography (MEG), Magnetic resonance imaging (MRI), diffusion tensor imaging (DTI), among others. In this session, we will present some cutting-edge development of statistics, aiming to address fundamental statistical problems arising from neuroimaging studies such as dynamic inverse of MEG image, non-normal frequency band analysis of brain signals and manifold learning on MRI and DTI data.