Harnessing the game-changing potential of emerging technologies is important in every sector – but within NHS Scotland it is absolutely vital.
Improving the quality, efficiency, and sustainability of healthcare must be a key driver in supporting the NHS Recovery Plan to underline and strengthen the innovation culture within our NHS.
That is where the new NHS Scotland Innovation Fellowship Scheme, launched in May, is aiming to make a significant contribution to a vibrant life science sector focussed on delivery to the NHS.
The scheme – which supports and invests in innovative opportunities developed through collaborations between NHS and Social Care, industry, and academia – has been honed by the Scottish Health and Industry Partnership (SHIP), a joint venture between the Chief Scientist Office and economy directorate within the Scottish Government.
Consisting of nine ground-breaking projects drawn from clinical staff across Scotland, the scheme’s first cohort of Fellows leading those projects is now in place.
It is firstly and fundamentally about taking great ideas, ranging from tech for preventing falls, machine learning-based tools to screen hip abnormalities in babies, and devices to support pre-diabetes detection in the pharmacy setting. It is then a case of fast tracking them into tangible solutions in order to address real problems in the NHS and social care. A further long-term ambition is encouraging NHS Scotland staff to establish innovation careers within their post.
The scheme will support fellows as they work with a multidisciplinary team to take forward innovation projects at a local and national level, and on completion of their Innovation Fellowship, will continue development of innovative solutions, building further capability and capacity.
One exciting Fellowship project has come from Professor Steven Lo, consultant plastic surgeon in the West of Scotland who has been focusing on a 3D telemedicine project – a world first for Scotland involving an international collaboration with Microsoft, Redmond, USA.
Professor Lo has utilised Microsoft’s Holoportation™ communication technology, a new type of 3D capture technology from Microsoft Research that allows high quality three-dimensional models of people to be reconstructed, compressed, and transmitted anywhere in the world in real time. When combined with mixed reality (MR) displays, the technology will allow users to see, hear, and interact with remote participants as if they are actually present in the same physical space. Communicating and interacting with remote users becomes as natural as face-to-face communication.
When used in medicine, Microsoft’s Holoportation™ communication technology can help doctors deliver a higher fidelity patient experience, with clinical trials indicating that this makes the clinical consultation more like an ‘in-person’ consultation with a doctor. With remote consultation now integrated into many clinical care pathways following the Covid pandemic, the need for more realistic forms of remote consultation is crucial. Microsoft’s Holoportation™ communication technology further allows clinicians to ‘draw’ operations virtually on 3D models of the patients, allowing the patient to clearly see and understand their operation, facilitating patient engagement and informed consent.
The development of the Holoportation™ communication technology has seen high patient satisfaction levels within NHS Greater Glasgow and Clyde, underlining the belief that its aim of closer face-to-face consultation is working. Indeed, 100 per cent of participants preferred the use of 3D for consultations compared to standard video calls.
Mixed Reality can open further doors to new ways of examination to support improvement in surgical planning, delivering successful interventions. Multiple team members can collaborate to determine the best solution for each patient and each procedure. Indeed, the use of Microsoft’s Holoportation™ communication technology has recently been used in Glasgow to deliver clinical consultations on complex reconstructive patients in Ghana, prior to an overseas surgical visit. This allows, for the first time, real-time assessment of complex reconstructive patients in 3D in lower to middle income countries – facilitating planning, international expertise and improving patient safety.
Another of the nine hugely promising fellowship projects involves the application of artificial intelligence (AI) in breast cancer detection, automating some of what a radiologist would do when they look at images of people who may have a disease, or do not have a disease, and decide what the next step is.
Dr Gerald Lip, clinical director for breast screening in the North East of Scotland and clinical lead of mammography work for the Industrial Centre for Artificial Intelligence Research in Digital Diagnostics (iCAIRD) – an AI centre of excellence based in Edinburgh and Glasgow – set out to evaluate a tool that would assist radiologists reading scans.
The project commenced in February 2019 with the essential aim of looking at ways in which AI could be applied to current breast screening practices to help radiologists achieve faster diagnosis and pick up early-stage cancers sooner, in turn allowing treatment to be targeted earlier.
With that aim in mind, Dr Lip and his team have been working with SME partner Kheiron Medical Technologies to evaluate Mammography Intelligent Assessment tool (MIA) for clinical-level performance and its capacity to meet the needs of the Scottish Breast Screening Programme through deployment across NHS Scotland.
Against a challenging backdrop of a shortage in radiologists, the potential to build a platform that can scale in future represents another exciting step.
If proven effective, the opportunity is clear – clinically, financially, and operationally – given that AI has the ability to read 200,000 mammograms every year in Scotland as a second reader.
Indeed, part of the project involves further scoping and preparation for the advanced development of MIA. This could mean implementation beyond iCAIRD and commercialisation.
The team is working with Canon, the University of Aberdeen, and NHS Grampian on the project’sthree main phases: retrospective evaluation of Kheirons’ MIA tool; real world evaluation in clinical practice in the Scottish Breast Screening Programme; and advanced development for further improving AI clinical decision support.
Encouraging progress was reported in November. It detailed how the retrospective evaluation’s data linkage had been performed with a paperless breast screening reporting system, allowing for complete anonymisation of data to facilitate external analysis.
The dataset comprises four years of consecutively acquired national screening activity drawn from a medium-sized centre.
The AI tool analysis was first applied to all images with a recommended threshold. Further analysis then followed using a standard machine learning (ML) format using a validation set and a final test set. No model alterations or re-training was conducted during or prior and after AI evaluation on this completely new data. The initial validation set comprised of 45,444 women and 303 cancers.
The human reader of the scans recorded a recall (identification of further tests being needed after an initial mammogram) rate of 5.74 per cent, while MIA recorded a rate of 12.97 per cent. Following that, the human reader’s cancer detection rate was found to be 5.38/1000 compared to MIA’s 6.1/1000.
Overall, it was concluded that AI implementation could mean possible workload savings of 30-40 per cent for mammographic reading, significantly freeing up clinicians.
Professor David Lowe, clinical director for health innovation at SHIP said: “The Innovation Fellowship Scheme is a key investment for the future of innovation within NHS Scotland. Cohort 1, primarily aimed at clinical NHS Scotland staff, has already been very well received. Future cohorts will be open to a wider group of applicants who demonstrate passion for innovation and the way it can contribute to improved health and social care delivery.
“Innovation has been identified as a key enabler of health and care recovery and re-design, and Scotland aims to be a global leader in its co-creation and adoption resulting in health, wellbeing, and economic benefit. Our triple helix approach brings together substantial expertise from the NHS, academia, and industry to facilitate creative solutions to our most pressing challenges.
“Going forward, the Innovation Fellowship Scheme will play a critical role in supporting Scottish Government efforts to radically transform health and social care delivery in ways that ensure sustainability and high quality, timely, accessible services. Embedded within this approach is enhancing productivity of the workforce during recovery and transformation of services.
“The talent and passion of the first cohort is evident, and we believe it is only the beginning as the scheme is foundational for accelerated innovation adoption.”
The other seven projects in Cohort 1 offer just as much exciting potential. In the North, alongside Dr Lip, is Andrew Radley’s Community Pharmacy Point-of-Care testing for early identification of individuals with pre-diabetes. In the same region is Priti Singh’s AI-assisted project which aims to develop innovative technological solutions that improve the digital connectivity and visibility of unscheduled mental health services across all tiers of health and social care.
Alongside Dr Lo in the West is Anna Talbot’s ‘Preventative/Proactive Technological Approaches to Frailty, Falls and Syncope’. This innovation fellowship will use digital technology to provide supportive self-management to help older people reduce their risk of future falls. This in turn will allow them to have fewer falls and spend more time in their own homes.
Also in the West, Chris Sainsbury is focussing on ‘Developing and applying machine learning based tools for clinical benefit in Diabetes, Endocrinology and beyond’ and is aiming to tackle the general problem of clinical data sharing using synthetic datasets.
In the South East meanwhile are three further fellowship schemes which hold huge promise.
Joyce Henderson is aiming to enhance AI-powered, portable diagnostic capabilities, improve patient outcomes, and develop efficient, effective clinical pathways through a focus on Developmental Dysplasia of the hip (DDH) surveillance.
It is an exciting development within NHS Scotland and a clear sign that the country is ready to be early adopters of the latest technical advancements.
AI within medical imaging is an exciting and fast-paced area of innovation and has the potential to revolutionise healthcare diagnostics. Rishi Ramaesh’s project is therefore looking to apply this learning for the earlier detection of lung cancer – in turn building on some of the existing collaboration between industry partners, academia, and the health service – and how we can use machine learning and radiomic analysis to personalise cancer treatments for patients.
Finally, Yvonne Chun is evaluating ‘Care Calendar’, an interactive digital innovation that is aiming to achieve high-quality and sustainable health care journeys.
Using the Care Calendar, clinicians can tailor care pathways to ensure that patients “receive the right care at the right time, every time.” Patients will be empowered in their own healthcare journey and clinical managers can utilise the data insights generated by Care Calendar to guide day-to-day running of the health service as well as service improvement.