Community Research and Development Information Service - CORDIS

H2020

Do CHANGE Report Summary

Project ID: 643735
Funded under: H2020-EU.3.1.

Periodic Reporting for period 2 - Do CHANGE (Do Cardiac Health: Advanced New Generation Ecosystem)

Reporting period: 2016-02-01 to 2017-01-31

Summary of the context and overall objectives of the project

Do CHANGE is a research & innovation project with European and Taiwanese partners, co-funded by the European Commission and the Taiwanese Government. It will last from early 2015 to early 2018. The primary objective of Do CHANGE is to develop a health ecosystem for integrated disease management of patients with ischemic heart disease and heart failure. The system will give them access to a set of personalized health services in a near real-time fashion. This disruptive system will incorporate the behaviour change method “Do Something Different”, in conjunction with new innovative wearable/portable tools that will give indications of salt and fluid intake, monitor behaviour and clinical parameters in normal living situations. The total ecosystem will provide guidance to the patients themselves, their families, informal carers, and patient's social environments for disease management. Also, clinical diagnosis by various health care professionals (dietician, psychologist, cardiologist, cardiology nurse practitioner, GP, case manager) will be supported, but only after they have been given authorised access to these data by the patients.

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

In the second year of the project, the second cycle of the co-design process has been completed. It was focussed on investigating the integration of Do CHANGE in the cardiac patient population via the privacy study, the app study, pre-EDL (Experiential Design Landscape) and the extended-EDL, which will give more insight in how to develop the responsive Do's and what the effects might be. From the patient interviews, different views on the topic of privacy within the health context have been revealed. This raises the challenge of flexibility within the design of a health ecosystem: it should be adoptable for different contexts and cultures and provide opportunities for different implementations.

During year 2 the innovative COOKiT sensor was developed based on chemical impedance. In the applied methodology, the sinusoidal signal is generated by the electronics which have been integrated in the COOKiT device. Many experiments were performed to create the analytical models and to test their repeatability, reliability and accuracy. Mid-fidelity prototypes of the FLUiT were included in a study with 40 renal patients. Fundamentals and measurement principles of the method developed for the liquids estimation have not changed in the final high fidelity version. Regarding complete meal scanning the development of the computer vision with the deep learning methodology of Convolutional Neural Network (CNN) was introduced for feasibility and affordability in the pilots. At the first stage of the development, manual calibration was used to improve the accuracy and to train the CNN module. A web application and smartphone application have been developed for the uploading and tagging of training data.

A new technical architecture consisting of distributed systems has been defined, with the high-level interfaces between them being specified and developed. The architecture will support the coordination of identity and privacy preferences between the distributed systems, such that information associated with an individual that is distributed between these systems can be shared in accordance with their preferences. The architecture will also support the sharing of pseudonymised information for individual analysis and “big data” analytics. A data model has been developed, based on data encapsulation using HL7 FHIR employing existing data coding schemes where available and a Do CHANGE project specific coding scheme where no suitable scheme exists. A data analytics framework has been set-up, which will be incorporated within the Do CHANGE architecture, processing collated data from the pseudonymised database with the aim of optimising the health outcomes of patients. It integrates relevant clinical data from some data sources, such as electronic health records and clinical notes, and non-clinical data obtained from mobile applications and questionnaires. For generating responsive Do’s the following parameters are calculated: those that reflect participants’ habits and behaviour in regards to their physical activity (Activity), the likelihood of engaging in social interactions (Social Opportunity) and the variety of experiences and activities in their days (Variety). After a baseline assessment, the system evaluates whether todays levels of Activity, Social Connectivity and Variety are above, equal to or below the participant average levels. When any of the parameters drop below baseline for several days, a “Responsive Do” is sent to encourage corrective action by the recipient.

The clinical evaluation has been split in Phase 1 and Phase 2 trials:
- Phase 1, planned to start on November 1, 2016, but effectively started per 15 Jan. 2017, both in Badalona and Tilburg;
- Phase 2, planned to start on 1 May 2017 in Badalona, 1 June in Dalin and Hsinchu, 1 July in Tilburg.
Both phases will continue for 6 months: each patient will get core Do’s during a 3-month period and then 3 months will be used to follow up the possible effects and sustainability of the targeted improvements. With the proposed two phases, in total 400 patients will be targeted to participate in the studies: 200 with interventions, 200 serving as control group.

Further investigation in future commercial exploitation of the research and innovation work from Do CHANGE suggests some novel commercially exploitable aspects not previously considered. It suggests different objectives and outcomes for the different client groups outlined. It also suggests that some of the main barriers to successful exploitation are inherent to the healthcare business sector – due to its very nature and purpose – and in some areas where exploitation is possible the issues to be overcome may require alternative ways of looking at the business options, such as social impact bonds.

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)

A new sensor has been developed to measure NaCl in a substance, based on chemical impedance. In the applied methodology, the sinusoidal signal is generated by the electronics which have been integrated in the COOKiT device. Regarding complete meal scanning the development of the computer vision with the deep learning methodology of Convolutional Neural Network (CNN) has been developed.
For combining data from different sources and granting acess to these data by the individual, a new technical architecture consisting of distributed systems has been defined, with the high-level interfaces between them being specified and developed. The architecture will support the coordination of identity and privacy preferences between the distributed systems, such that information associated with an individual that is distributed between these systems can be shared in accordance with the preferences of the individual.

The Do CHANGE ecosystem could have considerable impact economically, socially and medically if it can be deployed in part or in full in healthcare. The emphasis on behaviour change, however, is a fundamental element of the ecosystem. There are major human and economic benefits to be had in tackling the behavioural risk factors in disease and the Do CHANGE ecosystem could be used in many other health areas. There seems little doubt that many lives would be saved and healthcare costs reduced by changes in lifestyle of the kind being targeted in Do CHANGE.

Related information

Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top