Validation of 3D Simulations in Embolic Stroke

NCT05055960 | Unknown

• 1 other identifier: 0839
• Study Type: observational
• Target: 100
• Locations: 1 country
• Sites: 1

Brief Summary

Stroke is a common condition which results in significant disability for patients. There are different causes of stroke, but around one quarter are as a result of clots or other material from the heart lodging in blood vessels in the brain, stopping the blood supply to that area. Atrial fibrillation is a common cause of blood clots which go to the brain and can be easily treated with blood thinning medications, which significantly reduce the risk of further strokes. However, at the moment, atrial fibrillation is difficult to identify, and heart monitoring can be needed for up to one year. This significantly delays starting blood thinning medications and leaves patients at risk of stroke during this time. Therefore, better ways of picking up strokes caused by atrial fibrillation are needed. One such method may be to use brain scans which are routinely taken at the time a patient presents with an acute stroke. By using mathematical models to work out the source of stroke, we may be able to determine which strokes are caused by atrial fibrillation at the time the patient presents with their stroke. This would reduce the number of investigations patients under-go, saving money for the NHS, and reducing the number of tests patients have. Therefore, the aim of this project is to create an anonymised database of brain scans from patients who have presented to hospital with a stroke to develop and test these recently developed models to see if they can accurately identify which strokes are caused by atrial fibrillation, and which ones are not. This project has the potential to improve patient outcomes by reducing treatment delays and improving the accuracy of the diagnosis of the stroke source.

Conditions

Stroke

Outcome Measures

Primary Outcomes (1)

• A database of anonymous magnetic resonance imaging scans from patients presenting with acute ischaemic stroke for the development and validation of stroke simulation models.

  anonymised database

  1 year

Secondary Outcomes (1)

• The sensitivity and specificity of 3D and artificial intelligence models for determining the embolic source of the stroke against the current clinical reference standards

  3 years

Interventions

Stroke simulation and machine learning DIAGNOSTIC_TEST

Using simulation and machine learning approaches to classify the source of embolic stroke.

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

Adult patients presenting with acute ischaemic stroke to an acute hospital service.

You may qualify if:

• Patients presenting with acute ischemic stroke to the emergency department, transient ischaemic attack clinic, or hyperacute stroke unit
• Adults aged over 18 years
• Patients receiving a magnetic resonance imaging (MRI) scan

You may not qualify if:

• Insufficient clinical demographic and outcome data (\<75% of all variables)
• No acute stroke diagnosis or identifiable lesion on MRI
• Patients with a computed tomography scan only

Sponsors & Collaborators

• University of Leicester: lead
• Open University: collaborator

Study Sites (1)

University of Leicester

Leicester, Leicestershire, LE2 7LX, United Kingdom

Location

MeSH Terms

Conditions

Stroke

Interventions

Machine Learning

Condition Hierarchy (Ancestors)

Cerebrovascular Disorders; Brain Diseases; Central Nervous System Diseases; Nervous System Diseases; Vascular Diseases; Cardiovascular Diseases

Intervention Hierarchy (Ancestors)

Artificial Intelligence; Algorithms; Mathematical Concepts

Study Design

• Study Type: observational
• Observational Model: OTHER
• Time Perspective: RETROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Study Record Dates

• First Submitted: September 14, 2021
• First Posted: September 24, 2021
• Study Start: November 12, 2021
• Primary Completion: September 1, 2024
• Study Completion: September 1, 2024
• Last Updated: May 10, 2023

Record last verified: 2023-05

Data Sharing

• IPD Sharing: Will not share

Locations

GB (1)