Medipex NHS Innovation Awards 2022
For 2022, entries were invited from NHS-led teams from all over the UK, across four categories. Our panel of internal and external judges selected five finalists from each category, which were showcased at the awards event before announcing the winners of each category. The range and quality of innovations entered for the awards was inspirational again this year and it is anticipated that many new collaborations, and opportunities for shared learning, will occur as a result of the event. Congratulations to all of the finalists and winners, and thank you to all of our entrants for taking part.
We would like to give special thanks to the judging panel, made up of experts from Medipex Ltd., Medovate Ltd, the Yorkshire and Humber Academic Health Science Network and, the NIHR Children and Young People Medtech Co-operative and the NIHR Surgical Medtech co-operative, who had a very challenging time in selecting the finalists from a pool of nearly 50 high-quality entries from 14 different organisations and their collaborators.
Using Artificial Intelligence (AI) to improve patient services and/or safety
Innovations that are designed to use Artificial Intelligence (AI) to improve services (including hospital processes, diagnosis, screening, treatment and therapy) and/or improved patient safety.
Category winner
AI segmentation of cardiac MRI to automate the measurement of cardiac function and volume
Dr Andy J Swift, Dr Pete Metherall, Dr Kavita Karunasagaraar, Dr Samer Alabed, Sheffield Teaching Hospitals NHS Foundation Trust
When doctors ask for you to have a scan of your heart called magnetic resonance imaging (MRI), they usually want to understand how well your heart is pumping (function). This information can help your doctor to check your heart’s health, give you a diagnosis, and start or adjust your treatment. Measuring your heart function is a time-consuming task (delays of hours to days in the past).At Sheffield, a team of expert cardiac imaging specialists has worked with world-leading computer scientists to tackle this issue. They trained the computer to recognise the heart on MRI pictures and automatically measure its function. After years of research, the readings of the computer have achieved levels comparable with medical experts. The main advantages are that the measurements of your heart are immediately available. Getting answers quickly and accurately may cut the delay in receiving your right treatment.
Finalist
ANCHOR: an Artificial Neural network for Chronic subdural HematOma Referral outcome prediction
K. Joshi George, Salford Royal Hospital, Northern Care Alliance Foundation Trust
Chronic subdural hematoma (CSDH) incidence and referral rates are rapidly increasing. Many referrals are inappropriate and utilise significant clinical time for both the referring doctor as well as the neurosurgical team. Furthermore, acceptance of a CSDH referral is a subjective clinical decision influenced by several baseline prognostic variables. Accurate and automated evidence-based referral decision support tools are thus required to prevent unnecessary referrals and unwarranted interventions. Thus, we aimed to create a reliable machine learning (ML) algorithm capable of safely and accurately predicting the acceptance of a CSDH referral. Our best performing artificial neural network model is publicly deployed at https://medmlanalytics.com/neural-analysis-model/.
Finalist
CBCT based synthetic CT generation for adaptive radiotherapy planning
Christopher O’Hara, Leeds Teaching Hospitals NHS Trust
Radiotherapy uses high-energy radiation to treat cancer. It is likely that the patients shape will change during their treatment period and require their initial treatment plan, based on a CT scan, to be adapted. Cone Beam CTs (CBCTs), a less detailed CT scan, are taken during the patient’s treatment but do not have the required detail to make a conclusive decision on whether the patient’s treatment needs to be adapted. Synthetic-CTs, an image with CT-like quality, created from a CBCT is a solution to these issues. This work tested four methods of producing sCTs for patients with head and neck cancer and found that a deep learning model was the best clinical method. We are now assessing the accuracy of the deep learning model on different sites and developing a workflow that integrates this within an automated pathway for assessing whether a treatment requires adaption.
Finalist
Stratified Care guided by AI improves depression treatment outcomes
Dr Jaime Delgadillo, Rotherham, Doncaster and South Humber NHS Trust (RDaSH)
In England, around a million patients are referred each year to psychological services that treat depression and anxiety. These services offer low intensity treatments that only last a few weeks, and high intensity psychotherapies for those who need lengthier interventions. A longstanding challenge in this area is to decide which treatment to provide to each patient. We developed an AI technology based on principles of stratified medicine, which aims to offer ‘the right treatment, to the right patient, at the right time’. The technology enables mental health practitioners to decide whether patients should access low or high intensity therapies, using a machine learning algorithm that was developed by analysing data from over 1000 NHS patients. After co-developing this AI technology with industry partners, we conducted the world’s first clinical trial of AI-driven stratified care. Evidence from this study shows that this method improves depression treatment outcomes compared to usual care.
Finalist
Diagnostic AI System for Acute and Emergency Care (DAISY)
Dr Ol’Tunde Ashaolu, York and Scarborough Teaching Hospitals NHS Foundation Trust
Patients regularly wait a long time to access care in the Emergency Departments (ED). The wait is bound to get longer in these current national and global circumstances thus putting patients at risk. To tackle long wait times, strain on healthcare staffing and poor health outcomes, I have developed a complex algorithm that allows a (humanoid) robot (called DAISY) to triage patients in the emergency department. DAISY will greet patients who are prompted to input their symptoms whilst the algorithm uses this information as well as vital signs to diagnose and propose a treatment plan. This diagnosis and treatment plan are then reviewed and signed off by a doctor. We are currently testing DAISY before we move forward to a clinical trial to gather clinical data and patient and public opinions
Improved treatments, therapies and rehabilitation
Innovations designed to improve treatments for patients, (including, but not limited to, treatments involving surgery and devices to enhance surgical outcomes), and improved rehabilitation interventions.
Category winner
N-ICE Cream: High protein ice cream for hospitalised malnourished older adults
Professor Opinder Sahota, Nottingham University Hospitals NHS Trust
Poor nutritional intake is common in older people admitted to hospital, which is associated with significant food waste. These patients are routinely offered a nutritional supplement drink, but these are poorly consumed. Patients enjoy ice cream and although this provides calories, nutritionally they are very inadequate, with only ½ g of protein per serving. Recognising this preference for ice cream, in collaboration with DreamyCow dairy farm, we have developed a high protein vanilla ice cream, providing 19g of protein per serving with added vitamin D. In addition to providing the calories, the added benefit of the high protein and vitamin D helps reduces muscle loss. A taste preference study in 100 patients showed that ¾ preferred the N-ICE Cream nutritional supplement compared to the nutritional supplement drink. The expected benefits will see patients recover more quickly, reduce hospital length of stay, and provide cost savings by reducing food waste.
Finalist
CFHealthHub Learning Health system and habit lab
Martin Wildman, Sheffield Teaching Hospitals NHS Foundation Trust
The CFHealthHub Learning Health system and habit lab brings together an engaged community of practice in which multidisciplinary teams from more than 50% of English adult Cystic Fibrosis (CF) centres work with people with CF to support habits of preventative self-care and run evaluations and improvement programmes to optimise long term care delivery. The programme uses automatically collected data presented to people with CF and clinicians on a co-produced platform available on via the web and smartphones. The platform was evaluated in the largest CF randomised trial ever carried out in UK (19 centres 604 participants) and demonstrated significantly increased rates of preventative self-care associated with increased habit and reduced burden. The learning health system provides a trials within cohorts platform currently being used to carry out the largest global health technology assessment of new CF drugs (NEEMO) and understand equity via NICE indicator and just in time drug delivery (EMBRACE).
Finalist
Mindfulness for tinnitus, 8 week course delivered virtually
Mr P Arullendran, South Tyneside & Sunderland NHS Foundation Trust
Standard therapy for tinnitus commonly include the use of devices and relaxation exercises both of which give limited benefit. When NICE approved the use of mindfulness in 2022 Mr P Arullendran took the opportunity to look at areas within his clinical practice to introduce this service, with the restrictions imposed by COVID-19. All previous Mindfulness therapy published in the literature was delivered face to face. Work began with a focus group, which demonstrated effectiveness with an 8 week course when delivered online. When the results were audited the results were comparable to the improvements demonstrated in face to face courses. After the success of the pilot study the department agreed to fully fund a regular program for delivery of this service online
Finalist
Long covid holistic virtual rehabilitation programme
Rachel Tarrant and Jenna Shardha, Leeds Community Healthcare NHS Trust
In the service’s infancy, 1:1 therapy was provided to patients. With growing demand outstripping capacity, along with improved learning about the condition two virtual courses were set up to provide rehabilitation on a larger scale to groups of patients. One course managing fatigue and another for breathlessness. We found that a significant number of patients were needing to attend both courses. Given further learning about symptomology and presentation of Long Covid, one holistic programme encompassing all the main symptoms was created. This is now the service’s core intervention providing patients with education, tools and strategies to help self-manage their Long Covid symptoms. As well as increasing efficiency by delivering rehabilitation on a larger scale, it also provides peer support which is extremely valuable to the patient group. The virtual programme has been evaluated by our in-house researchers and has been further adapted in response to the feedback and emerging evidence.
Finalist
‘Let’s Go!’ Smartwatch App to support self-management and monitoring of paediatric incontinence
Nathaniel Mills, Sheffield Children’s NHS Foundation Trust
‘Let’s Go!’ is a smartwatch-based app designed with and for children with paediatric continence challenges, and their caregivers. It aims to empower primary-school aged children to self-manage and monitor their continence challenges in an asset-based way with their caregivers, particularly when they are at school. Caregivers can set discreet alarm reminders on the smartwatch, with customisable encouraging messages, to remind the child to drink and/or to visit the toilet. The child is able to snooze the alarm, document the action they have taken (drink, ‘wee’ or ‘poo’) and record an accident. Gamification elements (i.e. collecting badges) reward the child’s engagement with the app, rather than their current continence. ‘Let’s Go!’ is intended to complement, and be used alongside, clinical interventions. A working prototype is available for download and pilot testing. Expected benefits include increased understanding of, and confidence in self-managing, continence challenges, as well as accurate monitoring data to share with clinicians.
Delivering benefits through diagnosis and screening
Innovations designed to improve the detection and diagnosis of health conditions or disease.
Category winner
Novel salivary collection device for neonates, infants and young children
Dr Charlotte Elder, Sheffield Children’s NHS Foundation Trust
The use of salivary samples to screen and diagnose disease is growing in popularity. Salivary steroid measurements (cortisol) are already recommended in the diagnosis of hormone disease (Cushing Syndrome).The non-invasive nature of salivary collection, negating the need for blood tests and needles, makes it a particularly attractive medium for use in children. Currently there are significant obstacles to salivary collection in children under the age of six years, due to the requirement for active patient participation and the potential for choking. Traditionally cotton containing materials were used however these have since been shown to significantly compromise salivary cortisol analysis and their use is no longer favoured. Our SaliPac device, combining a pacifier (dummy) with a commercially available synthetic swab designed for salivary cortisol collection, provides a non-invasive salivary collection technique for steroid analysis in young children which has been shown to be effective and well tolerated in our studies
Finalist
Implementing a novel virtual reporting measure for detection of surgical wound infection
Dr Ross Lathan, Hull University Teaching Hospitals NHS Trust
Why we are doing the work – We identified that postoperative follow-up is not routinely undertaken and remote follow-up, where a patient is reviewed from their own home might be accurate to identify wound infections but the best way of doing so has not been investigated. What we are doing – We plan to combine two forms of remote follow-up, questionnaires and photography to see if this improves the accuracy of detecting wound infection compared to either of these methods alone. This involves the Bluebelle Wound Healing questionnaire, already used to assess infection and wound images taken by patients. The combined new method is ready for use but needs to be validated. Expected Benefits – The results of this project will enable patients to communicate with their surgeon from their own home, removing the need to travel to hospital unnecessarily. As such, it will remove any associated costs that would otherwise be incurred.
Finalist
Integrating microbiome analysis into NHS Bowel Cancer Screening to improve accuracy
Dr Caroline Young, Leeds Teaching Hospitals NHS Trust/University of Leeds
The current NHS Bowel Cancer Screening Programme (NHSBCSP) aims to detect bowel cancer before symptoms develop, by testing for blood in stool. If blood is detected, a camera test (colonoscopy) is offered. Unfortunately accuracy is limited; half of screening-colonoscopies are unnecessary. We are testing whether information about a person’s microbiome (bacteria in their stool) could improve screening accuracy. We have profiled the microbiome of 2252 routine NHSBCSP samples (gFOBT cards); results indicate that microbiome data has the potential improve accuracy. We are currently repeating the experiment using the new NHSBCSP screening device, the FIT test. We then plan to optimise the test, calculate how much money it would save, and determine the best way to roll it out to national screening. We hope that improving accuracy would reduce the number of unnecessary screening-colonoscopies, meaning that fewer participants would undergo this procedure without cause, reduce NHS costs and save resources.
Finalist
Foot pressure measurement quality assurance device
Chris Pearce, Sheffield Teaching Hospitals NHS Foundation Trust
Foot pressure clinics assess foot function, with measurements made of foot pressure in standing and walking trials. A foot pressure mat is used to produce pressure maps, indicating areas of high and low pressures, which are then used to inform clinical decision making. The accuracy and linearity of pressure mat measurements are very important and should be assured. Currently, other than an un-standardised spot check, assurance can only be achieved by sending the device to the manufacturer for calibration; this is costly and disruptive to clinics. We have designed and manufactured a standardised Quality Assurance device that applies a known pressure to regions of the mat, establishing its accuracy and linearity. Baseline measurements reveal variation across the pressure mat, which may be relevant clinically and demonstrates the need for such checks to be made regularly.
Finalist
Nurse-led lung cancer symptom assessment (LUMEN) service
Dr Rachel Gemine, Hywel Dda University Health Board
The Lumen Project is a Nurse-Led Lung Cancer Telephone Symptom Assessment Service, where Patients can self-refer via a dedicated telephone line. Patients, aged 40 or over, will have a telephone consultation with Clinical Nurse Specialist, patients meeting NICE Guidelines will be referred for chest x-ray. Rates of diagnosis of lung cancer have been drastically impacted by COVID-19. Our aim is to increase early diagnosis in lung cancer and improve patient experience and outcomes. We are evaluating the service, with the aim to develop a business case to increase service to cover the whole of the Hywel Dda University Health Board area and across Wales
Improved processes and systems
Innovations that are designed to improve efficiencies and outcomes for service delivery organisations.
Category winner
Consent for Contact (C4C)
Sarah Cooper and Crystal Romain-Hooper, Leeds and York Partnership NHS Foundation Trust
We are creating a list of service users who would be willing for our research team to contact them about active research opportunities relating to their current health care pathways. The clinical care teams are currently asking service users whether they would like to be contacted about research ongoing in the trust; and whether the service user consents or not to the C4C register is noted within their electronic medical records. If a service user agrees to join the C4C register they are giving consent for our research team to check their electronic medical records to determine eligibility to studies prior to directly contacting them. When approached about any specific study they will then be able to agree or not to participate. This register enables us to approach patients directly, increasing service user access to opportunities to take part in quality research projects and contribute to improve understanding of mental health conditions and wellbeing
Category winner
The Development of eTAROT (Electronic Transplant Assessment & Relative Opportunity Tool)
Pamela Hughes, Leeds Teaching Hospitals NHS Trust
The Electronic Transplant Assessment and Relative Opportunity Tool (eTAROT) is a laboratory based innovation developed by the Transplant Immunology Laboratory at Leeds in collaboration with the Renal MDT. It is intended to improve the likelihood of transplant for all wait-listed patients especially those who are immunologically complex. It achieves this by firstly applying a risk stratification system which ranks both national and locally developed renal transplant policies in increasing order of immunological risk and then by progressing patients through this system based on their length of waittime. In this way it balances risk/benefit with aspects of equality/equity of access to transplantation.
Finalist
Assisting Bradford and Craven beating Diabetes (AssistBCD) – a unique interactive clinical tool
Maria Pedley, Bradford District and Craven Health and Care Partnership
AssistBCD is a clinical decision-making tool designed locally with multi-disciplinary collaboration to support healthcare professionals to provide evidence-based care to individuals at high risk or living with Diabetes. The tool is embedded in SystmOne used by all practices and diabetes specialist teams within the Bradford district and Craven area. It has been designed to reduce variability, increasing parity and equity in service delivery across the district, to standardise and support evidenced-based decision making. The aim is to promote prevention of diabetes and improve clinical outcomes and quality of life for those living with diabetes.
Finalist
The Mid Yorkshire NHS Trust medical student and FY1 preparedness innovation programme
Dr Reshad Khodabocus, Mid Yorkshire NHS Trust
The Undergraduate Team have worked tirelessly with Clinical teams and Senior management to greatly increase face-to-face high-quality teaching to 650 medical students annually, providing them with exceptional clinical experience directly involved and supervised delivering high-quality patient care. We accelerated this programme during the first wave of covid-19; utilising students supporting the frontline workforce and have continued to move students to increasingly involved clinical roles supported by robust clinical supervision and pastoral support. They now have a huge range and quantity of interactive teaching delivered by the Undergraduate team centrally to better train medical students, make them more useful to clinical teams hosting them (over the 24/7 nature of the NHS Service work) and improve their lives as doctors. We also introduced 24 annual full-day themed FY1(First year medical doctor) training days, providing every FY1 with six highly rated training days that help them be the best FY1 they can be.
Finalist
Digital transformation of nursing documentation at Leeds Teaching Hospitals NHS Trust
David Pickles, Leeds Teaching Hospitals NHS Trust
This innovation will transform the 200+ different paper care plan and assessment documents nurses at Leeds Teaching Hospitals currently use. Working in partnership with Elsevier, the team will embed Elsevier’s evidence-based, peer-reviewed digital care plans into PPM+, the Trust’s electronic health record system, which is designed and managed in house by a dedicated team. This partnership allows the project to be completed rapidly and at scale. Led by nurses, and supported by the Trust’s IT colleagues, the new solution has the potential to significantly change the way nurses work, enabling quicker and easier decisions about patient care, simplifying daily work and saving time. It will allow Leeds to collect information about the impact nursing care has on patients to improve reporting, with a live feed of acuity, metrics and ward assurance data. Following comprehensive user research, the project is currently in the planning stages and will be launched early 2023.
Medipex CEO Lindsay Georgopoulos says, “Many congratulations to all of our finalists and of course our winners, who have provided much inspiration to our audience today. A special congratulations must go to our prize winners Professor Opinder Sahota from Nottingham University Hospitals NHS Foundation Trust, with his innovation N-ICE Cream, an innovative approach to reduce malnourishment amongst vulnerable hospital patients, and to Dr Andy Swift and team at Sheffield Teaching Hospitals NHS Foundation Trust, whose innovation AI segmentation of Cardiac MRI was particularly commended by the judges for the clear articulation of the need and benefits of their innovation.”