AI technology speeds up stroke care and reduces costs

AI technology speeds up stroke care and reduces costs

New artificial intelligence (AI)-driven imaging support software which has the potential to deliver huge benefits to hundreds of stroke patients each year and significant cost savings to the NHS has been implemented at the Royal Berkshire NHS Foundation Trust (RBH). The Oxford AHSN is supporting wider regional roll-out backed by NHS England. The CE-marked e-Stroke Suite developed by Brainomix Ltd includes tools for non-contrast CT and CT angiography brain scans, connecting physicians and facilitating information transfer. The software supports physicians in identifying patients who would benefit from acute stroke treatments, selecting patients most likely to benefit from specific interventions, such as mechanical thrombectomy.

Find your nearest AHSN

Coordinates
By Brainomix

Overview

Brainomix is an Oxford-based start-up which launched in 2010. Their vision is to “be a world leader in imaging software for neurological and cerebrovascular diseases”. The e-Stroke Suite software has been implemented in hospitals in Europe, Asia, North and South America.

The Oxford AHSN, Brainomix and RBH are partners in the National Consortium of Intelligent Medical Imaging (NCIMI), a network of NHS hospitals, clinical leaders, industry experts in the fields of AI and medical imaging, world-leading academic researchers plus patient groups and charities. NCIMI was formed in 2019 following a successful bid for Industrial Strategy Challenge Funding to drive innovation in the UK’s AI sector. The Oxford AHSN is supporting the roll out of the e-Stroke Suite across Thames Valley and the South East region.

In addition, with the Covid-19 outbreak, stroke teams that have local outbreaks have found themselves unable to provide a local stroke service with patients going to neighbouring units and non-stroke specialist areas instead, emphasising the importance of appropriate image-sharing software. The e-Stroke Suite supports rapid sharing of images across a network and new ways of working.

Stroke affects 110,000 people each year in the UK and it is the fourth largest cause of death. Around two-thirds of stroke survivors leave hospital with a disability. As a result, stroke costs UK society £26 billion per year (Source: Stroke Association). This number is expected to rise to between £61bn and £91bn by 2035 if no action is taken. The most severe and disabling strokes are due to a large vessel occlusion (LVO). Early identification and treatment of a stroke is of paramount importance.

The standard of care for LVO stroke in the UK has been administration of thrombolysis (a clot-busting drug called alteplase). This is only modestly effective for stroke patients with LVO. In 2015 a series of research trials demonstrated the efficacy of mechanical thrombectomy to open the occluded vessel for LVO stroke patients. This has transformed the potential care of patients with LVO stroke: patients who would have ended up in nursing homes are walking out of hospital the next day. Each thrombectomy is thought to save the economy £80,000 (reduced inpatient stay, reduced rehabilitation stay and less impact on the UK’s social care facilities after discharge). Source: Stroke Association

In the UK fewer than 1% of stroke patients undergo thrombectomy, but it is estimated that over 11-12% would benefit. The key barrier to delivering this treatment is patient identification due to a lack of real-time neuroradiology imaging expertise at all sites to which stroke patients present, compounded by communication barriers across stroke networks.

Within the Thames Valley, Oxford University Hospitals NHS Foundation Trust (OUH) provides specialist stroke care for patients requiring mechanical thrombectomy. In Oxford, approximately 30-50 patients undergo thrombectomy each year. The estimated number of patients across the Thames Valley who would benefit from thrombectomy is 300-400 per year.

How did the Innovation Exchange help?

The Oxford AHSN, Royal Berkshire NHS Foundation Trust and Brainomix are partners in the National Consortium of Intelligent Medical Imaging (NCIMI). This NCIMI exemplar project is led by  Brainomix and Dr Kiruba Nagaratnam. The aim of the project is to evaluate the feasibility of an artificial intelligence-enabled imaging support solution (e-Stroke Suite) combined with digital connectivity to improve stroke care in an NHS network.

Dr Guy Rooney, the Oxford AHSN Medical Director, has provided leadership and organisational support to the Thames Valley stroke network, chairing the Thrombectomy Innovation & Transformation (TITAN) quality improvement team who are leading on the roll-out of the e-Stroke Suite across the region. Five more hospitals in the TITAN network are due to come on line.

If the implementation is successful across a network, valued and adopted by clinicians and workflow is improved, a larger scale health economic analysis of the e-Stroke Suite will take place across several NHS stroke networks. Specific outcomes include:

  1. Demonstrating that an AI-enabled support solution can be integrated into an NHS network
  2. Impact on the total number of mechanical thrombectomy procedures and thrombolysis given across the region
  3. Evaluation of patient flow metrics including CT-to-decision time, door-in-door-out time in referring hospitals, door-to-needle time, number of patients transferred and proportion transferred who undergo mechanical thrombectomy
  4. Clinician satisfaction and practice impact evaluation at both referring and accepting hospital sites
  5. Cost-effectiveness analysis to contribute to a NICE technology appraisal.

The Oxford AHSN Strategic and Industry Partnerships team will support the evaluation of a six-month pilot to scale the solution across the NHS England South East region.

Impact & Outcomes

Fast and accurate decision-making is essential in ensuring that a patient suffering from an LVO stroke is diagnosed and transferred for thrombectomy rapidly. If LVO stroke is not diagnosed and treated quickly, the ramifications for the patient are enormous, often resulting in a severe disability. A core reason for this problem can be attributed to the fact that many patients eligible for thrombectomy are not receiving this treatment.

There are three factors to consider. Firstly, speed and accuracy are of the utmost importance when diagnosing patients with LVO so they can be referred on for treatment. Secondly, if this process is not carried out quickly the benefits of treatment drop off sharply, causing harm to patients and creating further costs to the NHS. Finally, specialists find it difficult to access CT scan images quickly in order to make these rapid, accurate diagnoses (particularly out of hours).

The e-Stroke Suite software from Brainomix addresses all three of these factors by providing a fast, easily accessible and accurate method of analysing CT scans for proximal vessel occlusions. According to the project lead at RBH, Dr Kiruba Nagaratnam, there are two situations where the benefits are particularly noticeable:

  1. When the image generated by the e-Stroke Suite is looked at by a non-specialist, the software will draw their eyes to where the problem is. So, instead of trying to get the result reported by a radiologist, the non-specialist can diagnose proximal vessel occlusion with confidence and the patient can be referred on. This reduces the time between the patient arriving in hospital and being referred on to treatment (where they are eligible).
  2. In cases where there is still uncertainty about a potential diagnosis, Brainomix imaging support software enables stroke specialists to analyse the CT scans accurately wherever they are. Prior to the implementation of this software, CT scans had to be reviewed by a specialist either within hospital or at home on a laptop with a secure wifi connection. Images can now be seen within two minutes of them being processed and advice given immediately. This again, reduces the time taken for an accurate diagnosis to be made, therefore making sure treatment is provided promptly and also provides specialists with a far easier and faster way of accessing the images given by a CT scan.

It is predicted that by implementing the e-Stroke Suite the RBH can increase the thrombectomy referral rate from 50% to up to 75% of eligible patients within existing working hours. By extending referrals to a 24/7 service 100% of patients would benefit. This not only increases the number of patients who benefit from treatment, but also increases the benefit the treatment can provide by enabling diagnoses to be made faster. Furthermore, length of stay in hospital is seven days less for patients who receive treatment for LVO stroke compared to those who do not.