European Commission logo
français français
CORDIS - Résultats de la recherche de l’UE
CORDIS

A patient-centered early risk prediction, prevention, and intervention platform to support the continuum of care in coronary artery disease (CAD) using eHealth and artificial intelligence

Periodic Reporting for period 1 - TIMELY (A patient-centered early risk prediction, prevention, and intervention platform to support the continuum of care in coronary artery disease (CAD) using eHealth and artificial intelligence)

Période du rapport: 2021-01-01 au 2022-06-30

Coronary artery disease (CAD) remains the leading cause of disease burden globally. CAD develops slowly, usually over decades, and depends on multiple (often modifiable) risk factors and their interactions also with individual disease predisposition. While the disease-centered evidence-based pharmacological and invasive interventions for CAD are well implemented at the clinical end, patient’s self-care assistance for disease control and prevention of CAD progression is highly fragmented across the European Union if implemented at all.
TIMELY is a platform that provides Artificial Intelligence (AI)-powered dashboards and decision support tools assisting patients and clinicians to personalize healthcare based on risk evaluation, outcome prediction and target interventions. TIMELY provides patient-centered eHealth tools (apps, IoT medical devices, communication), for patient empowerment and guidance during behavioral change interventions and healthy living, to manage their condition. The platform will be developed based on a CE-certified integrative database for Interoperability with electronic health records and security mechanisms, will ensure information completeness and continuity and will simplify data sharing. In any case, the TIMELY eHealth platform will be administered by a case manager supported by Artificial Intelligence, that will constantly monitor and evaluate risks and will indicate any deviation from defined therapy goals or unfavorable changes.
TIMELY will be developed as a modular and flexible platform that can be applied to provide health service for CAD patients across Europe also in countries with fragmented health service availability. The platform will be developed together with the most vulnerable and sensitive (in terms of technical assistance) group of CAD patients, i.e. those at higher risk due to ageing and comorbidities in the setting of inpatient cardiac rehabilitation. Behavioral scientists and medical/clinical psychologists will lead this task in an open innovation, living lab condition.
We have installed a living lab, i.e. an open innovation ecosystem centred on CAD patients and identified user requirements and needs for the TIMELY platform. Within the living lab a rich high-quality data lake is being generated for Ai studies with data of 300 CAD patients that are followed for 6 months.
Meanwhile the AI methods and models for the automated evaluation of patients’ risks (MACCE with machine learning, arrythmias with ECG analysis) as well as their expected response to cardiac rehabilitation programs are being developed in parallel with retrospective datasets (SERGAS, LURIC, GETABI, H-ADL, VIVIT) with promising accuracies, above 80%.
As part of the TIMELY platform, patients are provided with AI-powered apps. These apps serve three main functions. First, these apps act as diary-like assessment tools. Second, coaching app will provide motivating messages (or ‘nudges’), that are based on a range of validated behaviour change techniques. AI will be used to deliver these messages in a personalized and context-dependent manner. Finally, a chatbot app is being developed that is able to engage with brief motivating conversations, allowing a more in-depth assessment of barriers to chance lifestyle, and providing counselling interventions that draw from multiple therapeutic frameworks and approaches, such as motivational interviewing, self-determination theory, and cognitive behavioural therapy.
Key to secondary prevention the increase of cardiovascular fitness. Based on machine and reinforcement learning methods, another app is being developed that dynamically adapts the dosage and intensity of exercise training.
The core of the TIMELY-Net platform was planned and is currently being developed. It consists of a central back-end server, handling all the communication to AI microservices, import of measurement data and exchange with the patient portal, which is realized as mobile app for Android phones and iPhones. The Interfaces (clinicians’ dashboard and patient portal) are developed in parallel, as they are just two aspects of the same components.
Statistical analyses of retrospective datasets (TRIGGERS, THORESCI, Do-CHANGE) have resulted in recommendations for the development of the interventions. The behavioural change model was defined after a systematic literature review and focus groups with patients in Spain, Germany and the Netherlands. The clinical protocol has been revised and finalised and approval from local ethics committees is pending. MAST (Model for Assessment of Telemedicine) has been adopted as the method for the health technology assessment and data are being collected.
The communication strategy has been developed and is implemented. Dissemination activities, assets and tools have been developed and implemented: participation in 8 national and international congresses, preparation of papers for high-ranked journals (3 drafts), set-up of the Advisory Board and continuous exchange with it.
TIMELY will help to collect, integrate and interpret the large amount of already available patient-level data, also collected outside clinical and medical environments, and provide new streams of complex biomedical data for incorporation into practices and clinics to unfold the full potential of modern health care. TIMELY provides beyond state-of-the art continuous risk stratification, prediction of low adherence and/or effectiveness of the interventions and uses autonomous agents and AI chatbots to provide target interventions for optimizing CR core components.
TIMELY will conduct an RCT to provide best evidence for the effectiveness and cost-effectiveness of the proposed AI-powered, eHealth solution. TIMELY will also promote best practices and reduce variation of care delivery around Europe. TIMELY aims to provide better management of patients with diagnosis of coronary syndromes and comorbidities. The incremental value of using risk scores, such as Coropredict and AI-based, to evaluate patients' risks, and more importantly configure the eHealth components will be further evaluated within the project. TIMELY will offer effective secondary prevention services along the coronary heart disease pathway based on teaching self-management, adopting healthy behaviors including regular exercise, controlling biomedical indices and adhering to cardioprotective medicines. Patient activation and health literacy can relieve health systems from the cardiac disease epidemic.
For the care systems TIMELY can have an impact by facilitating and increasing referrals and enrolment in CR, as well as by increasing participation in it. In terms of secondary prevention TIMELY can lead to less readmissions, reduced recurrent cardiovascular events and reduced mortality rate. Which in turn will produce societal benefits such as increased productivity, reduced absenteeism from work and social events, reduced burden for caregivers and healthcare providers, cost efficacy compared to current care, better management of healthcare resources and reduced costs for medication and/or invasive procedures.
logo of TOMELY project

Documents connexes