In the first 18 months of the project, we focused on building a strong foundation for the RE-SAMPLE platform and services in terms of a collective knowledge base and active end-user involvement, leading to definition of our functional specifications, platform architecture, release of our first functional prototype, and the start of an observational cohort study for RWD gathering.
In months 19-36, we focused on implementing the RE-SAMPLE features and models, defining our privacy and security measures, and building our final prototype. Furthermore, we focused on defining the contents of our Virtual Companionship programme (VCP) by defining the intervention and coaching strategies, as well as on the integration of the VCP as part of daily care and daily life by defining the integrated care protocols. Next, we started with the dissemination of project results to our stakeholders, by numerous scientific publications, workshops, and social media activities. Lastly, we shaped our business strategy using market analysis and are advancing our exploitation strategy.
In the last 18 months of the RE-SAMPLE project, we have accomplished the following main achievements to reach our project objectives.
- Our RE-SAMPLE platform was deployed and running in production in all three pilot sites since August 2024 till August 2025, continuously collecting the data and combined with the hospital and Healthentia data in the edge nodes. AI models were trained, and personalised risk predictions and AI-services for shared-care decision making were provided to the clinicians via the clinician dashboard. A comprehensive security and privacy-preserving framework was established and implemented that guided the system’s architecture and development throughout the project. Following a strict privacy- and security-by-design approach, the consortium updated organisational, legal, and technical requirements ensuring compliance with GDPR and ethical standards across diverse European healthcare contexts. We advanced privacy-preserving machine learning through the design and implementation of a federated learning prototype leveraging homomorphic encryption and differential privacy, effectively balancing data protection with computational efficiency.
- The virtual companionship programme (VCP) was integrated in daily care and evaluated with its stakeholders in terms of usability, acceptability and societal impact. The final VCP provides tailored care and support for patients with COPD and comorbidity by offering 1) a virtual companion for the patient for ongoing self-management support and personalised coaching, 2) a clinical dashboard where the healthcare professionals can assess data and predictions, do shared decision making and follow-up on the advice provided, and for use with shared-care facilities. The results indicate no significant change in flare-ups, but a positive patient engagement and significant behavioural changes. From a societal perspective, the VCP is profitable for the coalition of stakeholders. Our findings suggest that the VCP is both feasible and promising, though further analyses (e.g. on patient compliance, quality of life, healthcare utilisation) are necessary to fully evaluate its clinical and economic impact.
- We have ensured that our results, data, models, and software are openly available. Specific attention in this phase of the project was for the communication, dissemination and exploitation of the results, to ensure the uptake of our results also after our project’s end. There were numerous scientific publications, workshops, and social media activities and strategic focus on key exploitable results reflecting the need to align innovation management efforts with concrete exploitation pathways. We also ensured that our real-world data ecosystem (data and software) is available through open access, coupled with accessible guidelines.
