Temporal dynamics of brain connectivity from resting-state and task fMRI data

Ranga Deshpande Ph.D.
Northwestern University, Chicago

BrainMap Seminar Series, 09/03/2019

The seat of power of the human brain is its extensive interconnectedness. Functional MRI is a spatially and temporally low-dimensional measurement of brain activity, and there has been widespread interest in the study of brain networks using fMRI. However, the predominant approach has been to view an entire fMRI run as a single snapshot to study temporally static brain connectivity. In this talk, I will present the alternative of studying brain networks as changing over time to quantify variability of connectivity in resting-state fMRI data. With higher variability of dynamics being associated with better cognitive behaviors, I will present the concept of rigidity (lower variability) as a tool to identify biomarkers of mental disorders. Using dynamic connectivity to improve the performance of machine learning prediction will also be discussed. With task fMRI data, I will present the utility of using dynamic connectivity to study the evolution of brain networks over successive task blocks/events, with applications in the study of emotion regulation/dysregulation and top-down modulation circuits. Finally, I will switch to the topic of modeling neurovascular coupling using fMRI hemodynamic response function (HRF), which is relevant to connectivity modeling as a confound variable. I will also present the alternative view of using the HRF as a tool to study brain function.

Athinoula A. Martinos Center for Biomedical Imaging
https://www.martinos.org/
Follow @MGHMartinos on Facebook, Instagram + Twitter!