Summary
More patients are getting the right treatment in the right place more quickly leading to better outcomes following a stroke thanks to a four-year collaborative programme led by Health Innovation Oxford and Thames Valley (Oxford HIN) supporting the deployment of artificial intelligence (AI) in stroke diagnosis. This approach has led to an increase in the number of stroke patients treated with mechanical thrombectomy (MT), a life-changing but time-critical intervention which reduces long-term disability following a stroke by reopening a blocked blood vessel in the brain. New image analysis technology enables high quality CT brain scans to be shared quickly and securely within and between hospitals using imaging software incorporating AI to support clinical decision-making. Working with the NHS England stroke teams and regional stroke networks, the Oxford HIN carried out the largest real world evaluation to date of AI in the NHS focusing on the impact of rolling out the Brainomix 360 Stroke AI tool. The evaluation was funded by the NHS Artificial Intelligence in Health and Care Award.
The evaluation found the treatment time was 50 minutes quicker and patients were 70% more likely to receive MT when AI was used to assist clinical decision-making.
This exemplar locality partnership demonstrates a breakthrough innovative approach, aligned with the prevention agenda and supporting economic growth. There is potential for further significant increases in MT rates enabling more people to regain independence following stroke.
Background
Around 100,000 people have a stroke in the UK every year and that number will increase by half in the next decade. Stroke is the leading cause of disability and exacerbates health inequalities. Early identification, rapid decision-making and prompt treatment can reduce brain damage and prevent or limit long-term disability. Mechanical thrombectomy (MT) has the potential to revolutionise stroke care and recovery. However, only 25% of eligible stroke patients are currently receiving MT. Timely referral to stroke centres with MT expertise is a crucial part of the treatment pathway. Imaging decision support software driven by AI can help to speed up clinical decision-making to determine patients’ suitability for treatment and transfer to their nearest thrombectomy centre. Most of the stroke centres in England and Wales have access to AI image interpretation. But evidence for using AI in the stroke pathway was lacking from larger real world studies.
First steps
Working with the NHS England stroke teams and regional stroke networks, the Oxford HIN carried out the largest assessment to date of the impact of rolling out new technology in the NHS with a detailed evaluation of the Brainomix 360 Stroke AI tool. The Oxford HIN initially facilitated strong collaboration through establishing the Thrombectomy Innovation and Transformation Network (TITAN) regional stroke network and introducing the AI tool into the Royal Berkshire NHS Foundation Trust in Reading as a ‘champion’ site through the clinical leadership of Dr Kiruba Nagaratnam. The Oxford HIN helped secure funding through the NIHR NHS AI in Health and Care Award which enabled scale up of the technology across a further four stroke networks across England and 25 stroke units. This funding also supported the Oxford HIN’s evaluation which included data from more than 83,000 patients obtained from the NHS national stroke audit (SSNAP).
Achievements
The evaluation found that where the Brainomix 360 Stroke AI image analysis technology was used to assist clinical decision-making, the treatment time was 50 minutes quicker and patients were 70% more likely to receive MT. In particular, sharing high quality CT brain scans quickly and securely is empowering non-specialists to refer with confidence. This exemplar locality partnership demonstrates a breakthrough innovative approach, aligned with the prevention agenda and supporting economic growth. The evaluation showed there is a major opportunity for systems to reduce the long-term burden of stroke care by significantly increasing access to MT.
This initiative also supported economic growth with Brainomix attributing the following to this programme of work:
- 29 jobs safeguarded
- 40 jobs created
- £400,000 investment leveraged
Feedback
“I think the major reason for the success in implementing AI in stroke is the network collaboration approach. We identified a clear need to increase the number of patients who receive mechanical thrombectomy. We had clear direction from NHS England policy on where AI can be used in the imaging pathway to improve detection of stroke. The stroke networks enabled deployment and scale up of the technology at pace. We engaged clinicians early on and were able to clearly demonstrate the benefits. We now have more confidence to support the use of AI in stroke detection – and we can now treat patients much faster.”
Dr Kiruba Nagaratnam, Clinical Lead, Buckinghamshire, Oxfordshire and Berkshire West Integrated Stroke Delivery Network, and Consultant Stroke Physician, Royal Berkshire NHS Foundation Trust
Kiruba discusses this project with Eleanor Wicks, a Consultant Cardiologist at Oxford University Hospitals, in this four minute video on improving outcomes following stroke recorded in July 2024 – see image above.
“The fantastic work of the stroke and interventional neuroradiology team at Oxford, and the wider TITaN network, demonstrates the power of a networked approach to complex medical interventions. The whole multidisciplinary team across the Buckinghamshire, Oxfordshire and Berkshire West Integrated Stroke Delivery Network should be congratulated for their phenomenal effort in delivering mechanical thrombectomy 24/7, reducing the severity of disability for many people.”
Dr David Hargroves, National Clinical Director for Stroke, NHS England
Next steps
There is potential for health systems to further extend access to thrombectomy and reduce the burden of long-term disability.