Data-driven approaches for identifying links between brain structure and function in health and disease

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Systems Neuroscience

Background:

• Brain imaging visualizes the living brain, offering insights into disease mechanisms and clinical decision-making.
• The complex relationship between brain structure and function is often overlooked in macroscopic, noninvasive studies.
• Understanding structure-function links is crucial for advancing neuroscience.

Purpose of the Study:

• To review data-driven approaches for linking macroscopic structural and functional Magnetic Resonance Imaging (MRI) data.
• To highlight the utility of independent component analysis (ICA) in exploring brain structure-function relationships.
• To demonstrate the potential of these methods in understanding mental illnesses, such as schizophrenia.

Main Methods:

• Focus on data-driven approaches, particularly independent component analysis (ICA).
• Leveraging higher-order statistics to integrate structural and functional MRI data at the macroscopic scale.
• Identifying covarying sets of brain regions and evaluating within- and between-network relationships.

Main Results:

• ICA enables the identification of structure-function links that may not be spatially congruent.
• These methods provide a network-level perspective, revealing coordinated activity across brain regions.
• Demonstrated potential in informing the understanding of mental disorders, with specific examples for schizophrenia.

Conclusions:

• Data-driven, network-level analysis of macroscopic MRI data is a powerful tool for understanding brain structure-function relationships.
• Independent component analysis offers unique advantages in uncovering complex, non-local associations.
• This approach holds significant promise for advancing our knowledge of neurological and psychiatric conditions.