NAL Interests ...
We are interested in the neuroimaging as a tool to understand neurodegenerative disorders including Alzheimer’s disease. We work on developing advanced analytics tools using state of the art machine learning/ deep learning and statistical methods to detect and predict pathological changes and ultimately guide treatment.
1.Highdimensional predictive modeling
We develop brain signatures using machine learning and high dimensional neuroimaging data to summarize the scans into a smaller number of makers reflecting various change such as age related atrophy or AD-like atrophy.
2.Pathological changes and lesion detection in MRI
Here we use specific deep learning architectures to detect small lesions such as cerebral micro-bleeds in the brain as well as larger lesions such as white matter hyper-intensities. Our goal is to understand factors associated with the appearance of those lesions.
The goal of this project is to understand the cause of death and guide pathological examination using postmortem MR scans. This projects gives uniques insights into evaluating radiological signatures with histopathology.
4.Data-driven disease subtyping
We are interested in developing new methods to disentangle heterogeneity and exploit the information from large scale cohort-based studies towards more specific subtyping for individuals. This research should have an impact on both understanding mechanisms and deliver efficient treatment.