Longitudinal predictive modeling of tau progression along the structural connectome

Area of Science:

• Neuroscience
• Medical Imaging
• Computational Biology

Background:

• Tau neurofibrillary tangles are key indicators of Alzheimer's disease (AD) progression and cognitive decline.
• Understanding tau transmission along neural pathways is crucial for predicting AD.
• Current methods lack personalized predictive capabilities for tau pathology.

Purpose of the Study:

• To develop an individualized predictive model for tau progression in Alzheimer's disease.
• To utilize a graph diffusion framework to model tau spread along the brain's structural connectome.
• To incorporate active tau generation and clearance into predictive modeling.

Main Methods:

• Developed an analytic graph diffusion framework with an inhomogeneous graph diffusion equation and source term.
• Utilized longitudinal tau positron emission tomography (PET) data (¹⁸F-Flortaucipir) and diffusion tensor imaging (DTI) from HABS and ADNI cohorts.
• Validated the model using two-timepoint PET scans for model fit and three-timepoint scans for predictive accuracy.

Main Results:

• The model demonstrated high consistency between predicted and observed regional tau measures.
• Analysis revealed differential tau spread patterns consistent with model predictions.
• Preliminary results show the model's potential for accurate in vivo tau progression forecasting.

Conclusions:

• The proposed graph diffusion model offers a promising personalized framework for predicting Alzheimer's disease tau progression.
• The model provides an in vivo macroscopic perspective on tau pathology spread.
• Further validation in larger cohorts is needed to confirm its prognostic value.