Patient Empowerment through Predictive PERsonalised decision support

Project Information

PEPPER

Grant agreement ID: 689810

DOI

10.3030/689810

Project closed

EC signature date 10 November 2015

Start date 1 February 2016

End date 31 March 2020

Funded under

• SOCIETAL CHALLENGES - Health, demographic change and well-being

Total cost

€ 3 548 071,25

EU contribution

€ 3 548 071,25

Coordinated by

OXFORD BROOKES UNIVERSITY
United Kingdom

Periodic Reporting for period 2 - PEPPER (Patient Empowerment through Predictive PERsonalised decision support)

Reporting period: 2017-08-01 to 2020-03-31

Summary of the context and overall objectives of the project

"According to the International Diabetes Federation, over 463 million adults were living with diabetes in 2019, and the associated annual health expenditure was at least $760 billion USD. Many such people regulate their condition by administering insulin several times a day, either by injection or insulin pump. PEPPER, short for Patient Empowerment through Predictive PERsonalised decision support, aims to empower these individuals to manage their condition more easily, by utilising portable technology, together with artificial intelligence (AI) and mathematical modelling, to offer freedom from daily decision-making.

Currently there is no decision support system for insulin dosing on the market that adapts itself based on real-time activity data and blood glucose data. The PEPPER system addresses this by providing personalised decision support on two alternative mobile platforms: one based on a smartphone, and another via the handset of a minimally-obstructive patch pump, which is about the size of a tic-tac box (Fig. 1). Users of the system also wear a fitness band and a continuous glucose monitor, which is around the size of a small USB stick. Additional information, such as carbohydrate consumption and alcohol intake, can be added manually on the handset (Fig. 2). Users’ safety is guaranteed through two levels of supervision. A first level consisting of a safety system including glucose alerts/alarm and constraints on insulin delivery, and a second layer consisting of a secure cloud-based server allowing remote supervision by clinicians.

Although PEPPER was designed to be used as a whole, it is also possible to use its individual components in an independent way. The corresponding PEPPER application program interfaces (APIs), have been produced by different partners within the PEPPER consortium. The approach used and resulting system architecture provide a generic framework for providing adaptive decision support anytime, anywhere, which could be applied to other health conditions that are monitored by wearable technology.

The system design process involved users at every stage to ensure that it meets patient needs and raises clinical outcomes as well as improving lifestyle, monitoring and quality of life. Prototypes of the system (Fig. 3) were demonstrated at various stages to a community of stakeholders including individuals with Type 1 Diabetes, the Vice-President of Diabetes UK, the Director of Research Partnerships from the Juvenile Diabetes Research Foundation (JDRF) and a representative of the Sociedad Española de Endocrinología y Nutrición. Tim Omer, representing the Nightscout (#WeAreNotWaiting) patient community, said ""As we capture higher quality and quantity of data about our condition, it is refreshing to finally see progress in assisting the patient with analysing this data to provide actionable feedback to reduce the burden of Type 1 Diabetes""."

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

The algorithmic work has focused on the development of an AI for decision support and the safety system, which uses a novel glucose forecasting algorithm. User needs were also carefully considered throughout the system design process.

The clinical validation phase began with a three-month feasibility study, which was divided into two parts: the first to test the safety system and the second tested the AI decision support system. Once all the data from the study were analysed, and parameters refined as required, the final prototype was released so that the concluding validation and safety study could begin. Usability evaluation also took place during all studies.

The results from the final validation study show that, despite limitations, the data is promising and suggestive that an adaptive bolus advisor, in combination with a safety system, may have potential to improve glycaemia and glycaemic variability. There is wide scope for integrating PEPPER into routine diabetes management for both insulin pump users and those on multiple-daily injections (MDI). Both, the adaptive bolus advisor and the safety system have potential for use within artificial pancreas systems.

The project team has built a strong associated research community, with highlights including prize-winning publications and joint organisation of three workshops on Artificial Intelligence for Diabetes, each co-located with a major conference, and involving editorship of associated journal special issues. Two project videos have been released in 2017 and 2018, garnering well over 7500 views between them. In addition, the number of publications in conference proceedings and journals is well in excess of the original project targets. The final wrap-up phase of the project focused on lessons learned, pre-commercial evaluation, and standardisation.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The PEPPER system goes beyond the state of the art by gathering data automatically from various sources using minimally invasive wearable technology integrated with manually entered data within a single platform. This facilitates a more accurate prediction of the effects of an insulin dose than previously possible, as well as improved, personalised adaptive decision support.


During pre-clinical testing, the industry standard UVA/Padova Type 1 Diabetes simulator showed improved results when testing each set of algorithms separately: both the adaptive decision support tool and the safety system, including predictive alarms, low-glucose insulin suspension, a novel carbohydrate recommendation, and a dynamic insulin constraint, perform better than preceding solutions. The clinical feasibility study of the latter also showed improved glycaemic control improved over the evaluated period. Percentage time in the hypoglycaemia range significantly decreased and there was a significant increase in percentage time in target range. There was also a reduction in the number of carbohydrate recommendations. The promising results of the final clinical validation study, despite being underpowered, show a clear trend toward improvement, and will be disseminated via a peer-reviewed paper submitted to the prestigious Diabetes Technology & Therapeutics journal.

The project has increased awareness of the potential of AI to be used in diabetes management, both among the scientific community and the general public. The project is also beginning to have some economic impact by enhancing innovation capacity of the commercial partner, Romsoft, which has reused tools from PEPPER in other projects of the company, thereby raising the level of competitiveness of these products. The APIs, released by four partners, including Romsoft, will enhance the exploitation potential of the algorithms. To date, three companies have expressed interest in commercialising some of the PEPPER modules. The outputs from the PEPPER system could have an enormous social impact since increasing the self-management and personal intervention of patients can only lead to better control of the disease and lower healthcare cost.

Screenshot of PEPPER Mobile Application

PEPPER Architecture

PEPPER phases

Cellnovo Pump System

PEPPER logo