The main aim of this study is to identify the predictive value of CT parameters (e-ASPECTS, CTP, collateral vessel status, volume, and location of ischemic lesion volume) at a 3-month functional outcome defined by the modified Rankin scale (mRS) in patients with non-lacunar stroke after recanalization treatment (Intravenous Thrombolytic Therapy and/or Mechanical Thrombectomy, or conservative treatment).

This is a multicenter observational study of patients with acute ischemic non-lacunar stroke and premorbid mRS ≤ 4, who are treated with intravenous thrombolysis and/or endovascular thrombectomy (EVT) or conservatively in stroke centers that are part of the National Stroke Research Network (STROCZECH).

All patients with suspected acute stroke routinely undergo baseline non-contrast computer tomography (NCCT) and single-phase CT-angiography (CTA) from the aortic arch to the vertex. Follow-up neuroimaging includes NCCT (a standard of care in the Czech Republic) up to 36 hours after endovascular therapy (EVT) or intravenous thrombolysis (IVT) treatment. The optimal timing of follow-up imaging > 24 hours is chosen since it represents the earliest time point for accurate delineation of acute ischemia volume.

NCCT will be performed on a multi-detector spiral 64 series CT machine. The NCCT examination is followed by CTA using 50-100 ml of iodine conjugate (Visipaque, GE Healthcare, Piscataway, NJ, USA), which is administered at a rate of 4 ml/s. The range of CTA is from the aortic arch to the distal intracranial artery. The width of the basic CT sections for further reconstruction is 0.75 mm.

Automated processing of NCCT, CTA, and CT perfusion (CTP) will be performed using the latest CE-marked version of e-Stroke software (Brainomix, Oxford, UK) at baseline, and follow-up imaging will be processed using algorithms in development by Brainomix.

The e-Stroke image processing algorithms follow an artificial intelligence (AI) approach with a combination of traditional 3D graphics and statistical methods, and machine learning classification techniques. The algorithms have been trained on a large dataset (> 10000 images) containing a wide range of real-world CT scans from stroke patients and negative controls with ground-truth data from additional imaging data such as MRI acquired within 1-2 hours of the CT scan, along with other modalities and clinical information. This dataset contains examples of CT scans captured with scanners from all major manufacturers, from a wide range of countries worldwide. Within e-Stroke NCCT, CTA and CTP will be processed using e-ASPECTS, e-CTA, and e-CTP modules respectively. It must be noted that e-Stroke is intended to be used as a decision support tool. Results are designed to be interpreted in the clinical context and radiological interpretation of the user.

Standardly acquired non-contrast CT, CTA, and CTP will be processed.

Within this project, imaging data from the e-STROKE system and clinical data stored in the RES-Q registry (REgistry of Stroke Care Quality) will be fused and subsequently analyzed. The RES-Q registry collects information related to hospital care for stroke patients, especially in the acute phase. Since 2021, the Ministry of Health of the Czech Republic has included RES-Q in the national policy as one of the indicators of stroke care. Therefore, the range of parameters assessed within this project will provide unique and potentially new information related to stroke that could impact the provision of non-hospital care in the future.

Inclusion Criteria:

• Acute ischemic stroke within 12 hours of onset, treated with intravenous thrombolysis and/or mechanical thrombectomy or conservatively.
• Unknown time of onset of ischemic stroke.
• Wake-up stroke (with symptoms of stroke since awakening).

Exclusion Criteria:

• Pre-event modified Rankin Scale (mRS) score > 4 points.
• Acute haemorrhagia or other NCCT finding excluding dg. ischemic stroke.

Consecutively admitted patients with ischemic stroke over the age of 18.

There is no plan to make individual participant data available to other researchers.