Multi-object Data Integration in the Study of Primary Progressive Aphasia

Area of Science:

• Neuroimaging
• Neuroscience
• Medical Imaging

Background:

• Primary progressive aphasia (PPA) is a neurodegenerative disorder impacting speech.
• Understanding the neurobiological underpinnings of PPA is crucial for diagnosis and treatment.
• Multi-modal imaging data, including gray matter (GM) structure and functional magnetic resonance imaging (fMRI)-derived brain connectomes, offer rich information.

Purpose of the Study:

• To identify specific brain regions associated with the speech rate measure in PPA patients.
• To develop an integrated framework for analyzing structural and connectivity data in PPA.
• To gain insights into the neuro-degenerative patterns of PPA.

Main Methods:

• Developed an integrated object response regression framework for network and GM images.
• Proposed a novel integrated prior formulation for network and structural image coefficients.
• Utilized a Bayesian framework to analyze multi-modal brain imaging data (GM and fMRI connectomes) and speech rate measures.

Main Results:

• The integrated framework successfully identified brain regions significantly related to the speech rate measure in PPA.
• The novel prior formulation effectively exploited brain connectome information and interconnections between GM and network data.
• The Bayesian approach provided characterized uncertainty in identifying brain regions linked to PPA.

Conclusions:

• The study offers new insights into the relationship between specific brain regions and PPA.
• The developed framework provides a deeper understanding of neuro-degenerative patterns in PPA.
• This multi-modal imaging approach advances the study of neurodegenerative disorders.