Advancing Neuroimaging Science

The Applied Neuroimaging Statistics Research Laboratory is an academic research group dedicated to improving statistical methodologies for analyzing functional magnetic resonance imaging (fMRI) data. Our research spans multiple domains, including the development of novel computational tools for studying the brain’s structural and functional connectome. We leverage state-of-the-art techniques such as multi-modal data fusion and machine learning to address fundamental questions in neuroimaging and to enhance our understanding of neurological and psychiatric disorders.

Our work bridges theory and application, combining methodological innovation with empirical investigations in neuroimaging. By integrating advanced statistical approaches with neurobiological data, we aim to refine how we study brain networks, mental illness, and cognitive function.

Development of Advanced Computational Methods for Neuroimaging Analysis

  • Innovate and apply machine learning and statistical methodologies, including tensor and matrix decomposition techniques, independent component analysis (ICA), linked ICA, and normative modeling, to analyze multimodal neuroimaging data.

Reproducibility and Generalizability in Neuroimaging Research

  • Ensure rigor in computational neuroimaging studies by adopting best practices for replicability, including ReproNim (Reproducible Neuroimaging) standards and large-sample validation methodologies.

Big Data and RDoC-Based Approaches to Psychiatric and Neurological Disorders

  • Utilize large-scale, open-access neuroimaging datasets to apply NIMH’s Research Domain Criteria (RDoC) framework for understanding neuropsychiatric conditions as variations along a normal-to-pathological continuum.
  • Map neurocircuitry associated with cognitive and emotional systems and study how deviations from normative models contribute to neuropsychiatric symptoms across aging and disease populations.

Understanding Neuropsychiatric Symptoms in Alzheimer’s Disease

  • Investigate how neurocircuitry degeneration in Alzheimer’s Disease (AD) relates to neuropsychiatric symptoms (NPS) such as depression, anxiety, agitation, and apathy.
  • Utilize the Triple Network Model (Default Mode Network, Central Executive Network, Salience Network) to examine mechanisms underlying NPS in AD.
  • Leverage large-scale datasets (HCP, ADNI, CRHD) to build normative models of brain-behavior relationships and identify deviations that contribute to disease pathology.