Cardiovascular disease (CVD) remains the leading challenge for European healthcare, affecting an estimated 85 million people across the continent. For decades, the medical community has relied on a “one-size-fits-all” model—treating patients based on population averages rather than individual biological truth. However, iCARE4CVD, a €22 million collaborative mission funded by iHi and involving 38 global partners, is working to end the era of medical guesswork.
At the heart of this shift is a simple idea: treating patients are not averages.
As iCARE4CVD coordinator Prof. Hans-Peter Brunner-La Rocca explains, personalised medicine makes it possible to diagnose earlier, predict risk more accurately, and choose treatments that truly fit each patient. Instead of adjusting therapies through trial and error, care can become more targeted, timely, and effective from the start.
But making this a reality is not just about better science—it’s about making it work in everyday practice. As Dr. Arantxa Barandiaran, cardiologist and iCARE4CVD coordination member, highlights, personalised medicine has clear potential to improve early detection, risk assessment, and treatment decisions. However, these benefits can only be realised if tools are integrated into clinical workflows. Without systems that are easy to use and fit into already busy routines, even the best innovations risk remaining unused.
The Expert Diagnostic: Identifying the Barriers
If the potential of personalised medicine is so clear, why does the “average patient” model still dominate?
The answer lies not in a single missing technology, but in a set of interconnected challenges that continue to slow real change. Across the iCARE4CVD consortium, three systemic barriers stand out.
1. The Data Paradox
Today, vast amounts of health data already exist—from hospitals and labs to wearables and home monitoring. But as Dr. Katja Zilonova – Senior Research Scientist at Novo Nordisk – explains, this data is scattered across systems that don’t communicate with each other.
The result? Even the most sophisticated AI models are built on small, fragmented datasets.
To unlock real value, this data needs to be connected and used at scale while still respecting privacy. As Dr. Zilonova argues, this requires that governments treat health data as national infrastructure – a “moonshot” effort akin to the space race—to unify records and streamline ethics.
2. The Interoperability Gap
Even when data is accessible, it often speaks different “languages. Different hospitals and systems record and structure information in different ways, making it hard to combine or compare. As Dr. Sathish Sankarpandi (CTO at Orbital/Vitturi) highlights, this lack of interoperability creates real barriers in clinical practice. Doctors are often left navigating fragmented systems instead of seeing a complete view of the patient.
On top of this, many datasets are not representative of all populations. This can reduce the accuracy of predictive models for certain groups, raising both technical and equity concerns.
3. The Reimbursement Barrier
Innovation is often hindered by outdated incentive and remuneration models. As Dr. Bettina Zippel-Schultz (Head of Innovation, German Foundation for the Chronically Ill) points out, current systems reward standardised, procedure-oriented care. Personalised, patient-centered approaches—especially those involving prevention and digital tools—are often inadequately covered by existing reimbursement structures.
At the same time, Dr. Benoît Tyl (Early Clinical Lead at Bayer) highlights a structural mismatch between the size of patient populations and the resources available to deliver personalised care. Current approaches prioritise broad interventions that provide average benefit, but do not distinguish between responders and non-responders, limiting both efficiency and outcomes.
The iCARE4CVD Solution: AI and Federated Data
This is where iCARE4CVD focuses its efforts: moving CVD care from “guessing” to “predicting”.
At the core of the initiative is a federated data approach, enabling researchers to connect existing health data from thousands of patients without requiring it to leave its original source. This allows large-scale analysis while preserving patient privacy—creating the foundation for more robust and clinically relevant insights.
Building on this infrastructure, we are developing AI models that move beyond general risk assessment towards identifying which patients are most likely to respond to specific interventions. This supports more targeted treatment decisions, improving outcomes while reducing unnecessary treatments.
At the same time, integrating imaging (MRI/CT), genomics, and blood-based biomarkers enables a more comprehensive understanding of disease, capturing individual variation and supporting more informed clinical decision-making.
Together, these advances enable a shift from population-based care to more precise, individualised treatment. To make this actionable in practice, digital tools integrated into electronic health records can support clinicians within their existing workflows—helping them make better decisions without increasing workload, while also supporting patient adherence.
From Possibility to Practice
Personalised cardiovascular care is no longer a question of scientific feasibility, but of implementation. Across the iCARE4CVD consortium, the challenge is clear: aligning data infrastructure, clinical workflows, evidence generation,
and reimbursement models to move beyond pilot innovation and into routine care.The opportunity is equally clear. Earlier intervention, more precise treatment, and better outcomes for millions of patients are within reach.
What remains is coordinated action—across research, healthcare systems, and policy—to ensure that personalised medicine delivers impact at scale.
“In clinical research, we often focus on medical outcomes. But for patients, quality of life, daily functioning, and personal priorities are just as important. If we want to make a real difference, we need to take these into account.”
