Accurate diagnosis of stroke due to large vessel occlusion (LVO) is an essential step in providing acute stroke care to a community. The gold-standard for diagnosis LVO is brain imaging, which is impractical in the pre-hospital setting. A non-invasive method to detect LVO is needed. Using highly sensitive accelerometers, one can measure the "HeadPulse"- tiny forces exerted on the skull from the cardiac contraction. This study tests the hypothesis that LVO alters the HeadPulse characteristically. Analysis of these data along with the subjects vascular status (LVO vs. non-LVO as measured by CT angiography) will be used to create a model that can predict LVO status in suspect stroke subjects.

Large vessel occlusion (LVO) ischemic stroke is a treatable disease at specialized centers. Clinical outcomes of patients with LVO is time-dependent. To maximize clinical benefit of thrombectomy, stroke systems of care need accurate methods to triage patients with LVO to comprehensive stroke centers or thrombectomy ready centers rather than closer primary stroke centers.

EPISODE will evaluate whether measurements of the HeadPulse can predict the presence or absence of LVO based on recordings from a consecutive series of patients undergoing CTA for suspected stroke. Recordings from patients pre- and post-thrombectomy will also be compared to investigate whether a systematic change in cranial forces exists during LVO.

Inclusion Criteria:

• Suspected acute stroke
• CTA or MRA performed

Exclusion Criteria:

• Any patient in whom obtaining a non-invasive recording is judged by the treating team to interfere with clinical assessment or treatment of the patient.

Patients who have been evaluated by a medical professional who feels the subject may be experiencing an acute stroke and has triggered the acute stroke algorithm within the medical center.