I have broad experience in statistical analysis, mathematics and computer programming. I am especially interested in fMRI experimental design, fMRI/DTI/PWI/PET image and signal processing and data analysis. In particular, fMRI experimental design for cognitive and psychology studies, data analysis methods development, and their application in clinical domains including dementia. For example current work includes:
This involves phase-encoded design, block design (random) and event-related design. Because fMRI is indirect measure of human brain activation, designing experimental approaches, which maximally induce brain activation, is vital for fMRI post-processing.
It is essential to develop sophisticated methods to exploit the information in the rich signal acquired from the brain activation. This is because the complexity of the brain signals is detected by the MRI/PET scanner, while being tolerant to noise sources at the same time is extremely difficult. I am especially interested in application of multi-level fMRI data analysis, DTI and ODF for Q-ball imaging regularisation and fiber tracking, absolute quantification of cerebral blood flow (CBF) using DSC-PWI imaging and VBM methods.
For qMRI, I have been undertaking T1 and proton density images estimations using flip angle method. I computed T2/T2* images based on regression methods and segmented brain tumor using various statistics methods. Recently, I involved in applying Bayesian theory and machine learing methods including deep learning methods for brain tumor classification.
Application of novel brain image processing techniques to study abnormal brain, such as brain tumour, Parkinson's and Alzheimer’s diseases and human amblyopia.