Connectome-based predictive modeling for functional recovery of acute ischemic stroke

Area of Science:

• Neuroscience
• Medical Imaging
• Neurology

Background:

• Acute ischemic stroke patients have significant recovery potential in early weeks.
• Functional recovery prognosis is crucial for post-stroke care and placement decisions.
• Traditional prognosis relies on demographics and clinical factors, but connectome analysis offers new insights.

Purpose of the Study:

• To develop connectome-based predictive models for post-stroke functional recovery.
• To forecast functional assessment scores using brain connectivity data at different time points.

Main Methods:

• Utilized resting-state functional MRI (fMRI) to compute brain connectivity.
• Developed predictive models using brain connectivity at stroke onset and one month post-stroke.
• Assessed functional recovery using modified Rankin Scale (mRS) and Barthel Index (BI).

Main Results:

• Significant models predicted mRS at one month and BI at three months post-stroke for supratentorial infarction patients using baseline connectivity.
• Predictive models also forecast mRS at three months post-stroke using one-month post-stroke connectivity for right hemisphere supratentorial infarction patients.

Conclusions:

• Connectome-based predictive modeling shows potential clinical value in prognosing acute ischemic stroke recovery.
• Brain connectivity patterns can forecast functional outcomes, improving patient care strategies.