My research uses ideas from machine learning and statistics to tackle some of the interesting/challenging problems in neuroscience. Here, I summarize some of my past and present research projects.

Joint Modeling of fMRI and DWI Data (Current)
    The interaction between functional and anatomical imaging modalities provides a rich framework for understanding the brain. In particular, correlations present in resting-state fMRI data are believed to reflect the intrinsic functional connectivity of the brain. Similarly, tractography based on DWI data is used to estimate underlying white matter fiber tracts (anatomical connectivity). To date, relatively little progress has been made in fusing information from these measures.
    We propose a novel probabilistic framework to infer the relationship between these modalities. The model is based on latent connectivities between brain regions and makes intuitive assumptions about the data generation process. Additionally, we formulate a natural extension of the model to population studies. [Paper]

Extracting Robust fMRI Measures for Clinical Diagnosis (Current)
    Pinpointing robust differences between a control and clinical population is crucial for understanding the effects of a disease. This task is especially difficult with fMRI data due to the considerable amount of noise/inter-subject variability and to the relatively small number of subjects recruited for clinical studies. Univariate tests and random effects analysis are the standard techniques employed in this application. However, this analysis ignores multivariate patterns within the data and may not be robust.
    Our approach is to apply feature selection/classification algorithms to identify preictive differences between the populations. We successfully employed the Gini Importance measure derived from Random Forests to identify functional connectivity anomalies induced by schizophrenia. We are currently looking at task fMRI for social anxiety disorder. [Paper]

Data-Driven Functional Connectivity Analysis (2008-2009)
    Functional connectivity analysis aims to detect and characterize coherent patterns of activity in resting-state fMRI as a means of identifying brain systems. Seed-based correlation analysis is the most popular technique. However, it requires a priori knowledge of the brain's functional organization and depends on consistent seed placement.
    We investigate the application of data-driven clustering algorithms as complementary approach to seed-based analysis. This allows us to partition the whole brain into an increasing number of clusters. We show that both K-Means and Spectral Clustering yield patterns that correspond to known functional systems. [Paper]