Computational Methods For Analysis Of Resting-State Functional Connectivity

Speaker: Harini Eavani , UPenn

Date: Wednesday, March 19, 2014

Time: 3:00 PM to 4:00 PM Note: all times are in the Eastern Time Zone

Public: Yes

Location: 32-G531

Event Type:

Room Description:

Host: Polina Golland

Contact: Polina Golland, x38005, polina@csail.mit.edu

Relevant URL:

Speaker URL: None

Speaker Photo:
None

Reminders to: seminars@csail.mit.edu, v-golland@csail.mit.edu

Reminder Subject: TALK: Computational Methods For Analysis Of Resting-State Functional Connectivity

The functional organization of the brain and its variability with development in youth can be studied using resting state functional MRI (rsfMRI). The spatio-temporal complexity of the connectome poses a challenge for conventional uni-variate analysis approaches. Motivated by models of neuronal activity which are driven by sparsity, we are working on incorporating recent advances in sparse representations for modeling spatial patterns of functional connectivity. To be able to resolve the patterns along time, we propose the use of state-space modeling techniques. Features extracted from these spatial and temporal patterns will be used for subsequent group analyses. Thus, these innovative modeling techniques address open research problems in each of three dimensions - spatial extent of functional systems, temporal changes and inter-subject variability. Its application to a large neuro-imaging dataset of over 1400 individuals will identify trajectories of normal functional brain development and serve as the baseline for other investigations into developmental disorders.

Research Areas:

Impact Areas:

See other events that are part of the Biomedical Imaging and Analysis 2013/2014.

Created by Polina Golland at Thursday, February 20, 2014 at 10:10 PM.