This study seeks to use safe, powerful, non-invasive computing tools, including machine learning and advanced neuroimaging analysis, to better understand how stroke affects the brain's network of connections. Using structural MRI, including diffusion-weighted imaging, this study will generate a detailed map of brain pathways to evaluate how strokes in the middle cerebral artery (MCA) territory disrupt the brain's structural networks. In the future, this approach may help physicians better predict recovery, monitor neuroplasticity, and guide rehabilitation decisions after stroke.

Stroke is one of the leading causes of long-term disability worldwide, with motor, cognitive, and functional impairments that often persist for months or years after the initial event. A central challenge in post-stroke care is the ability to predict individual recovery trajectories, which remain highly variable even among patients with similar clinical presentations. Traditional prognostic tools such as the National Institutes of Health Stroke Scale (NIHSS) and the modified Rankin Scale (mRS) offer population-level trends but are limited in their capacity to reflect the nuanced, network-level impact of focal brain injury.

Recent advances in neuroimaging and network neuroscience have shown that stroke is not solely a focal disease, but one that disrupts distributed brain networks. Lesions often disrupt not only local cortical and subcortical areas but also distant, structurally and functionally connected regions. This phenomenon, known as diaschisis, contributes to impairments that cannot be explained solely by the visible infarct. In addition, secondary degeneration, and the reorganization of brain networks over time play a significant role in shaping recovery trajectories. These insights suggest that understanding how a stroke alters the brain's connectivity patterns could offer new avenues for more precise and individualized prognostication.

Functional recovery is driven by preserved region reorganization and compensatory network recruitment. Previous studies have demonstrated that areas with greater structural and functional disconnection were more likely to undergo functional reorganization over time. Furthermore, the extent of early post-stroke reorganization was significantly correlated with long-term motor recovery at six months. These findings underscore the potential of connectome-based biomarkers to serve as early indicators of recovery potential and targets for rehabilitation planning. Notably, studies have shown that these network-level features differ between stroke subtypes and are correlated with clinical severity and outcome, supporting their potential role as biomarkers of recovery.

Despite these promising findings, connectomic methods remain underutilized in clinical settings due to technical complexity and the absence of standardized tools for interpretation. However, clinical platforms such as Omniscient's Quicktome now offer automated and anatomically informed visualization of structural and functional brain networks derived from standard DWI and rs-fMRI data. While these tools have been applied primarily in neurosurgical planning, their use in stroke prognostication is an emerging area of research.

There is a growing need to bridge the gap between clinical neurology and network neuroscience by validating connectome-based tools in the context of acute stroke care. Integrating connectomics with standard clinical assessments may improve the accuracy of outcome prediction, guide patient-specific rehabilitation strategies, and support the development of individualized recovery profiles.

The study will: 1) create a prospective, observational dataset to evaluate MRI-derived structural and functional connectivity changes in patients with distal middle cerebral artery (MCA) strokes, including M1 and more distal occlusions who have received mechanical thrombectomy and/or intravenous thrombolytics; 2) include patients with residual motor deficits in the acute setting following reperfusion therapy, while excluding those with completed M1 infarcts; 3)assess the feasibility and validity of using connectome-based metrics (e.g., tract integrity and disruption patterns) to quantify white matter connectivity patterns; 4) correlate connectivity patterns with motor outcomes at 3 months using the key clinical assessments; NIHSS motor , Modified Rankin Scale (mRS), DRAGON scores, and THRIVE scores ; and 5) evaluate whether acute-phase connectomic profiles can predict long-term functional outcomes and contribute to the development of a "recovery potential" scale.

Inclusion Criteria:

• Age > 18 years;
• Diagnosis of acute ischemic stroke with confirmed occlusion of the M1 or more distal MCA territory.
• Received reperfusion therapy via mechanical thrombectomy or IV thrombolytics(Tenecteplase, or Alteplase).
• Presence of a motor deficit on initial clinical exam (e.g., NIHSS > 0) and on immediate post-intervention exam.
• Patients or health-care proxy must be able to provide informed consent.
• Must be able to undergo sequential MRI at Lenox Hill Hospital, including resting-state fMRI (rs-fMRI) and diffusion MRI (dMRI) for, respectively, functional, and structural connectomic analyses.

Exclusion Criteria:

• Age < 17 years;
• Large vessel occlusions proximal to M1 (e.g., ICA), completed M1 occlusions.
• Pre-stroke Modified Rankin Scale score ≥ 3
• Known neurodegenerative disease or prior stroke affecting motor pathways.
• Inability to undergo MRI due to cardiac pacemaker, claustrophobia, and metal implants that cannot be removed prior to MRI.
• Pregnancy. Because of potential risk of serial MRI to fetus, women of child-bearing age require a pregnancy test at screening and agree to contraceptive practices during the study.
• Poor image quality or incomplete imaging datasets.