£1m project to test safety of AI in Scottish healthcare

Glasgow Times

3 days ago

Funded by Innovate UK, the scheme brings together hospitals, Glasgow and Edinburgh universities, and technology companies to create a validation framework for AI tools in the Scottish NHS.
The project involves collaboration between NHS Greater Glasgow and Clyde, NHS Lothian, and AI evaluation company Aival.
Read more: 'I would not be here today': Glasgow nurse's quick thinking saves colleagues life
Dr Rishi Ramaesh, consultant radiologist and innovation fellow at NHS Lothian, said: "Artificial intelligence has tremendous potential to improve patient care, but healthcare leaders need confidence that these systems are safe and effective.
"This project will help healthcare leaders to evaluate AI, and make sure that new technologies deliver real benefits for patients."
Aival's independent evaluation platform will be used to assess AI systems for diagnosing head trauma and lung cancer, aiming to improve care for patients and support NHS staff.
The platform allows hospitals to verify AI performance using anonymised patient data and provides ongoing monitoring once the software is deployed.
The project will also test the Aival platform's ability to monitor long-term AI performance, addressing concerns about 'drift'—the decline in software accuracy over time due to changes in patient populations, disease trends, or equipment updates.
Dr Mark Hall, consultant radiologist at NHS Greater Glasgow and Clyde, said: "Post-deployment surveillance monitoring is a critical yet often overlooked aspect of patient care, especially in radiology, where early detection of disease progression can make all the difference.
"Despite its importance, there are currently no standardised guidelines.
"AI-powered monitoring software bridges this gap by providing a structured approach."
One of the challenges addressed by the project is the lengthy testing process for AI.
Currently, it can take more than nine months to evaluate a single product, and there are more than 200 AI options available for some hospital departments.
This has limited the rollout of AI solutions in clinical settings.
The project will compare six commercial AI products used in stroke and lung cancer triage, including tools developed by InferVision, Annalise-AI, and Qure.AI.
Luciana D'Adderio, Edinburgh University academic and AI evaluation and assurance expert, said: "AI technology is achieving widespread deployment across healthcare settings, yet its assurance has not received the rigorous attention it demands.
"There is an urgent need for innovative tools and technologies for AI assurance, which themselves must undergo thorough evaluation and validation.
"This critical challenge forms the cornerstone of our groundbreaking project."
Kanwal Bhati, CEO of Aival (Image: Emelie Holgersson)
Read more: New general manager expected to be 'big hit' at Glasgow care home
Kanwal Bhatia, chief executive officer and founder of Aival, said: "It's vital that we monitor and check AI that's being used in decisions on patients' health, to ensure the best outcomes for patients.
"Any healthcare workers using AI need to be sure that the product is doing what it says it does – not just now, but five years in the future.
"Putting in place effective validation systems will encourage trust and adoption of AI, and will deliver cost savings and growth in the NHS and in private healthcare.
"We work hand in hand with NHS leaders, clinical and technical teams to provide the expertise and software to ensure that their AI systems are doing what they're supposed to do."