Advancing post-stroke outcome prediction with movement-specific structural and functional brain atlases.

Area of Science:

• Neuroimaging
• Stroke Rehabilitation
• Biomarker Discovery

Background:

• Stroke is a primary cause of death and disability, with motor deficits being a major contributor.
• Existing brain atlases for predicting post-stroke motor outcomes have limitations in coverage and integration of structural/functional data.

Purpose of the Study:

• To develop and evaluate novel neuroimaging biomarkers for predicting post-stroke motor outcomes.
• To construct comprehensive sensorimotor brain atlases integrating structural and functional data.

Main Methods:

• Developed a sensorimotor structural connectivity atlas (SSCA) and a probabilistic sensorimotor activation-based atlas (SMAA).
• Analyzed lesion load in 142 ischemic stroke patients and correlated it with Action Research Arm Test scores at 12 weeks using multivariable linear regression.
• Compared predictive performance against existing atlases and baseline Fugl-Meyer Assessment scores.

Main Results:

• The SSCA showed moderate predictive performance, but was outperformed by the Sensorimotor Area Tract Template.
• The SMAA achieved comparable predictive performance to other atlases with reduced model complexity.
• Atlas-based lesion load metrics correlated with upper limb motor outcomes but offered limited additional predictive value beyond baseline scores.

Conclusions:

• Novel sensorimotor brain atlases can predict post-stroke motor outcomes.
• The SMAA offers a more efficient approach compared to broader tract coverage atlases.
• Further refinement and multimodal approaches are needed for improved prediction of motor deficits.