Automated Tractography Settings for Assessing the Corticoreticular Pathway in Patients With Stroke: A Clinical Application

Background: The corticoreticular pathway (CRP) is gaining attention for its potential role in post-stroke motor recovery. However, unlike the corticospinal tract (CST), the CRP is not included in commonly used automated tractography frameworks, limiting its reproducibility and clinical applicability. We aimed to incorporate CRP assessment into an automated tractography pipeline and to evaluate its feasibility in patients with mild stroke. Methods: We retrospectively analyzed patients with a first-ever unilateral supratentorial ischemic stroke who had regained full independence in activities of daily living. Diffusion tensor imaging was acquired during the second week after admission using a 3.0-T MRI scanner. Tractography was performed using a fully automated procedure that reconstructed both the CST and the CRP. For CRP reconstruction, the seed region was placed in the gigantocellular reticular nucleus and the target region in the premotor cortex, based on established neuroanatomical atlases. Tract volume, fractional anisotropy (FA), and CST-CRP overlap were quantified. Lesioned and non-lesioned hemispheres were compared using the Wilcoxon signed-rank test. Results: Fifteen patients met these criteria during the study period. Automated tractography consistently reconstructed both pathways in all participants. CRP volume tended to exceed CST volume, although with substantial interindividual variability. FA values were relatively preserved in the lesioned hemisphere but were significantly lower than in the non-lesioned hemisphere for both the CST (S = 48, P = 0.004) and the CRP (S = 54, P = 0.001). Conclusions: Using anatomically defined regions of interest enabled reliable automated CRP reconstruction. This approach offers a practical and reproducible method for evaluating motor-related white matter pathways and may facilitate future research into stroke rehabilitation.