Fully exploratory network ICA (FENICA) on resting-state fMRI data.

Area of Science:

• Neuroimaging
• Data Analysis
• Cognitive Neuroscience

Background:

• Independent Component Analysis (ICA) is a key method for analyzing resting-state networks (RSNs) in fMRI.
• Current group-level ICA methods often require manual inspection or predefined templates for single-subject components.

Purpose of the Study:

• To apply FENICA, a novel group ICA method, to resting-state fMRI data.
• To assess FENICA's ability to identify spatially consistent RSNs across subjects without manual or template-based selection.

Main Methods:

• Applied FENICA to resting-state fMRI data from 28 healthy subjects.
• FENICA relies solely on spatial consistency across subjects to identify networks.

Main Results:

• Identified eight distinct group-level RSNs.
• The identified RSNs corresponded to known networks: visual, default mode, sensorimotor, dorsolateral prefrontal, temporal prefrontal, basal ganglia, auditory, and working memory networks.

Conclusions:

• FENICA provides a truly explorative approach for assessing RSNs.
• This method automates the identification of spatially consistent networks, eliminating the need for manual or template-based component selection.