Articles

Marilena Bicchieri is a Project Manager at Humanitas Research Hospital, where she oversees the institution’s participation in European and international projects. Her scientific background combines a degree in chemistry with a PhD in molecular oncology, and a research focus mostly dedicated towards investigating the therapeutic use of microRNAs for breast tumours.

As part of our series of interviews with GenoMed4All partners, Marilena reflects on the role she’s fulfilled during the project and her hopes for its impact on advancing precision medicine for hematology after the project ends in June 2025.

Work package leadership and interdisciplinary collaborations

In GenoMed4All, I’ve been working as part of the team responsible for use case validation. Our focus was to assess the feasibility and clinical value of the GenoMed4All platform and its AI-driven tools in real-world clinical settings. This meant piloting the platform across various institutions and seeing how it held up, not just in terms of data sharing through federated learning, but also how applicable and effective the AI models were in supporting clinicians through different phases of the patient journey, from diagnosis to prognosis and treatment decisions.

We addressed three specific use cases: sickle cell disease (SCD), myelodysplastic syndromes (MDS) and multiple myeloma (MM). Each came with very different clinical and technological challenges; for example, sickle cell disease is hereditary, whereas the other two are not. All three are rare diseases, so data scarcity was a recurring issue.

To tackle these complexities, we worked hand in hand with clinicians. We started by identifying the unmet clinical needs for each use case — such as diagnosis, risk prediction and treatment decisions — and built our methodology around those needs. A big part of our role was supporting the standardization of data collection across sites, drafting protocols, harmonizing data and defining clinical outcomes in collaboration with both clinicians and technical partners.

At Humanitas, we were especially focused on bridging the clinical and technical worlds. It was important to ensure that the AI solutions developed were actually usable and relevant in a clinical context. For instance, we put a lot of effort into creating a unified, FAIR-aligned methodology for data standardization. We also activated new collaborations with European registries like ENROL and RADeep. Ultimately, we evaluated the performance of the federated AI models through real pilot trials, always with an eye on their clinical utility.

Impact on precision medicine for hematology

What excites me most about the work we’ve done as part of GenoMed4All is that it shows how technological innovation can genuinely make an impact in clinical practice. This isn’t just a research exercise, we were able to demonstrate that AI models can be integrated into actual clinical workflows, offering tangible benefits for patients.

By using a federated learning approach and building a privacy-preserving infrastructure, we were able to bring data-driven, personalised care to the forefront while fully respecting patient data privacy. The tools we developed enable a much more tailored and effective way to support clinical decision-making, and I believe that this kind of integration is key for the future of precision medicine.

Learnings from this 4-year journey

One of the biggest takeaways for me has been the importance of a truly multidisciplinary approach. The project brought together a diverse range of stakeholders with different backgrounds, and that diversity was fundamental to its success.

Another major lesson is that technological innovation should always be guided by clinical needs, not the other way around. For AI tools to be trusted and adopted, clinicians need to be involved right from the start. At the same time, technical partners need to understand the real-world clinical challenges they’re aiming to solve.

The legacy of GenoMed4All going forward

Looking ahead, I believe GenoMed4All leaves behind a strong and lasting legacy. We’ve laid the foundation for a federated, privacy-preserving infrastructure that connects hospitals and research centres across borders. We’ve also built a tight-knit community of institutions, researchers and clinicians who are motivated to keep pushing this work forward.

Thanks to the support of ERN-EuroBloodNet and other key networks, we’re already thinking about how to expand. We want to bring more centres into this effort, and we’re putting a strategy in place to grow this community. Ultimately, GenoMed4All is not just a project — it’s the beginning of something much bigger.

Gastone Castellani is Full Professor of Physics and Biophysics at University of Bologna, where he combines his extensive background in physics of complex systems and biology. He is also a renowned figure in the field of Artificial Intelligence (AI) applied to medicine and healthcare, having spent 20 years at Brown University in the US during the 70s collaborating with other prominent AI researchers, including Leon Cooper and John Hopfield (both recipients of a Nobel Prize in Physics).

As part of our new series of interviews with GenoMed4All partners, Gastone reflects on the role he’s fulfilled during the project and his expectations for its impact on advancing precision medicine for hematology after the project ends in June 2025.

Work package leadership and interdisciplinary collaborations

In GenoMed4All, I’ve been leading Work Package 6, which is dedicated to developing artificial intelligence algorithms for the project. This work package has brought together a diverse and highly collaborative group, and our focus has been on designing AI models for both tabular data — like genomic information — and medical imaging.

We’ve worked across three use cases: sickle cell disease, multiple myeloma and myelodysplastic syndromes. Alongside our partners, we’ve developed a range of algorithms, and we’ve made our software openly available through a shared web repository for the benefit of GenoMed4All collaborators and the wider research community.

One of our most significant achievements has been the implementation of federated versions of these algorithms. Not all models are suitable for federated learning, but for those that are, we’ve successfully deployed and trained them within a federated network of GenoMed4All partners. This decentralized approach — where algorithms move instead of sensitive patient data — is one of the hallmarks of the project and a critical step forward in privacy-conscious AI development.

Impact on precision medicine for hematology

Precision medicine is an ambitious and evolving field, and while no single project can address all its challenges, I believe our contributions offer tangible progress.

For example, in the context of Sickle Cell Disease, which can be considered as a relatively underexplored condition, we developed a novel image analysis algorithm to detect “silent infarcts,” or white matter hyperintensities, using MRI data. These are very small lesions that are often hard to detect and easy to confuse with other brain regions. Our model, validated on over 500 MRI scans, has been well received by neuroradiologists, and we’re now submitting our results to a major journal.

In the area of Myelodysplastic syndromes, we’ve collaborated closely with other GenoMed4All partners to identify a genomic signature that can predict the risk of progression to acute myeloid leukemia, marking an important clinical milestone.

And in the case of Multiple Myeloma, we’ve combined genomic data with imaging techniques like PET and CT scans. This integration has led to more accurate predictions of overall and leukemia-free survival, and we’re currently preparing a publication on these findings.

It’s this kind of multimodal, integrative approach that I believe holds great promise for the future of personalized medicine.

The legacy of GenoMed4All going forward

One of the biggest challenges we faced was accessing data. In my experience, this is a recurring issue in many EU projects: it often takes two years or more — nearly half the project’s lifetime — to secure the data necessary for analysis.

That said, I’m hopeful that one of the lasting legacies of GenoMed4All will be the federated infrastructure we’ve established, which should significantly reduce these delays in future projects. This setup allows us to train models where the data resides, without needing to physically move sensitive information, something that aligns well with modern data governance principles.

Another important takeaway is that the approaches we’ve developed for hematological diseases can be extended to other areas of medicine, such as lung cancer, diabetes and neurodegenerative disorders. Hematology, much like neuroscience, is at the frontier of precision medicine, and the tools we’ve built here are widely applicable beyond the original scope of the project.

Looking ahead, I’m also excited about emerging fields such as synthetic data generation and digital twins. Through initiatives like the EU-funded project SYNTHEMA, we’re already exploring how to generate and validate synthetic datasets, which could provide an alternative when real patient data is hard to access; and digital twin technologies could enable us to model individual patient trajectories in-silico, not just in hematology, but across multiple disease areas. Thanks to GenoMed4All, we already have the federated software and infrastructure in place to support these next steps.

Silvia Uribe Mayoral is an Associate Professor at Universidad Politécnica de Madrid (UPM), where she’s been involved in various EU-funded projects focused on ICT applied to healthcare innovation, including GenoMed4All and SYNTHEMA among many others. With an academic and professional background in Telecommunications Engineering, she also holds a PhD in Technology and Communication Systems.

As part of our new series of interviews with GenoMed4All partners, Silvia reflects on the role she’s fulfilled during the project, and her expectations for its impact on fostering Federated Learning systems in clinical contexts after the project ends in June 2025.

Work package leadership and interdisciplinary collaborations

At UPM, my team and I have been leading efforts within Work Package 4, where our primary focus has been the development of the Federated Learning Platform. Our goal has been to address one of the key challenges in clinical AI: how to access and utilize sensitive clinical data for algorithm development while respecting the strict privacy requirements inherent in healthcare.

We began by engaging closely with various stakeholders, both from the clinical side and the software development community, to understand their specific needs. Based on this input, we designed and implemented a fully operational federated learning platform. 

This platform enables data scientists and other users to develop AI algorithms directly on distributed datasets across different clinical nodes, without the need to centralize the data; in this architecture, the data stays where it is, and only the algorithms travel. This approach significantly enhances data privacy and compliance, especially in a clinical environment where patient confidentiality is paramount.

Impact on precision medicine for hematology

I believe our contribution is particularly important for the advancement of precision medicine in hematology. 

By allowing stakeholders to access and analyze distributed clinical data without physically moving it, our platform opens up new possibilities for developing AI-driven diagnostic and decision-support tools. It respects privacy regulations while maximizing the value of data, which is crucial when working with highly sensitive and personal information.

The legacy of GenoMed4All going forward

Reflecting on my experience with GenoMed4All, one key takeaway is that despite all our advancements, there are still many unmet needs within the healthcare system. Technology, especially in the field of AI, has tremendous potential to address these gaps, but it requires a deep understanding of the context and a commitment to designing solutions that are both effective and ethically sound.

Looking ahead, I see the federated learning platform we’ve developed as a strong foundation for future work; it’s already been deployed across several nodes and has demonstrated real functionality. I hope it will serve as a springboard for new initiatives, helping us to tackle further challenges in clinical AI and continue building solutions that meet the evolving needs of healthcare.

Anna Rizzo is a Research Project Manager at Datawizard in Italy. Her current expertise in dissemination and exploitation for EU-funded projects, particularly related to e-health, draws from a combined academic and professional background in biology, data science and science journalism.

As part of our new series of interviews with GenoMed4All partners, Anna reflects on the role she’s had during the project, her contributions and main learnings, and her expectations for the project’s impact after it’s officially completed in June 2025.

Work package leadership and interdisciplinary collaborations

As a healthcare IT company with a strong focus on telemedicine applications and data standardisation, at Datawizard we’ve had the privilege of contributing to both the early and final phases of the project.

Our involvement began with data modelling, gathering requirements from clinicians and technical partners and translating those needs into functional and technical specifications for the GenoMed4All platform. One of the core challenges was selecting a coherent data model that could support the complex needs of the project, alongside harmonising and standardising diverse data sources.

We also worked closely with the clinical partners to support the characterisation of data flows across the project; this was essential for ensuring ethical and legal compliance, particularly under the GDPR, and towards the final phase we’ve also been focused on aligning our work with emerging frameworks like the EU AI Act and the technical standards being developed by bodies such as CEN-CENELEC.

Another key contribution was supporting the exploitation strategy. At Datawizard, we operate at the intersection of research and market implementation. We’re not only committed to advancing innovation at the European level but also to turning those innovations into usable products for hospitals and industry, particularly in telemedicine. In GenoMed4All, we took on the role of bridging the gap between research and the market; we’ve worked with partners to review project outcomes, identify potential use cases, and support efforts to give those results a life beyond the project — whether through further research, new funding opportunities, or early-stage market pathways.

Impact on precision medicine for hematology

One of the most common challenges with EU-funded research is that excellent work often remains confined within project boundaries. We’ve tried to counter that by helping bring GenoMed4All innovations closer to real-world application — or, as we like to say, closer to the patient’s bedside.

One major advantage has been our strong connection to ERN-EuroBloodNet, the European Reference Network for Rare Hematological Diseases. Through this collaboration, we’ve been able to ensure wider dissemination, support training programmes and, most importantly, lay the groundwork for the GenoMed4All platform’s sustainability.

Two new networks are currently being developed as part of the project’s legacy. The first, led by Vall d’Hebron Research Institute, focuses on non-oncological hematological conditions like Sickle Cell Disease; the second, led by Humanitas Research Hospital, is dedicated to hematological cancers. These networks will act as central nodes, connecting clinical centres with academic and private-sector partners who are developing AI-driven decision support tools.

The goal is to enable those partners to refine and validate their solutions using high-quality clinical data from the network. If we succeed, this will represent a major step forward in advancing precision medicine in hematology.

The legacy of GenoMed4All going forward

From the beginning, we understood that the GenoMed4All platform wasn’t meant to become a commercial product. Its value lies in its structure, its collaborative ecosystem, and the quality of its data. That’s why we’ve agreed within the consortium that the platform’s long-term stewardship should rest with ERN-EuroBloodNet and the clinical centres involved. They are best positioned to ensure continued research, support solution development, and train the next generation of tools and professionals.

This handover represents both an honour and a responsibility for the clinical centres. But with the support of the consortium and the broader ERN-EuroBloodNet community, it also marks the beginning of a federated research network in clinical AI. That, to me, is one of GenoMed4All’s most important achievements.

Main takeaways from this 4-year journey

There are always lessons to be learned from working on exploitation. One recurring challenge is getting partners, especially those from academia, to engage with the idea of exploitation. Many are focused on publishing results and advancing research, which is entirely valid, but they often have little experience or comfort with the concept of valorising those outcomes beyond academia.

This disconnect extends to relationships between academia, clinical centres, and industry. These groups often operate with very different goals and expectations. But for exploitation to succeed, we need to understand who the future users and buyers of our solutions might be — whether hospitals, companies or funding agencies.

Even when solutions are still at an early technology readiness level, we should think ahead: What does success look like beyond the lab or the pilot? Who might benefit from this innovation, and how can we bring them into the conversation early? That kind of forward-thinking is essential if we want our research to have a real impact, and I hope it’s a mindset we continue to cultivate across EU-funded projects.

Nathan Lea is the Information Governance Lead at The European Institute for Innovation through Health Data (i~HD). Drawing from his 25+ years of experience in Digital Health, backed by a masters in Computer Science and a PhD in Information Governance and Security, Nathan has become a widely renowned figure in regulatory affairs for clinical data and patients’ privacy rights.

As part of our new series of interviews with GenoMed4All partners, Nathan reflects on the role he’s had during the project, his contributions and main learnings, and his expectations for the project’s impact after it’s officially completed in June 2025.

Work package leadership and interdisciplinary collaborations

Over the past few years, I’ve been deeply involved in leading Work Package 2, looking after compliance, data governance and ethics. 

My main focus has been to help the consortium develop a complete understanding of what we’re trying to achieve, how we’re doing it, and why it matters. That starts with understanding how data flows through the project, but it goes far beyond that. It’s about building a shared sense of purpose, ensuring that everyone, regardless of their role, understands what’s being done, with what data, and toward what end.

In multi-partner projects like this one, it’s easy for teams to get siloed. But GenoMed4All really brought together people who might never have collaborated otherwise, and we managed to create something cohesive, something greater than the sum of its parts. Even those not directly involved in data governance came to appreciate the importance of, for example, how informed consent at the beginning of a process affects tool development further down the line.

That kind of mutual understanding doesn’t come from compliance checklists or rigid directives. It comes from dialogue. From saying, “Let’s figure this out together.” That’s the approach we’ve taken throughout WP2 — not as enforcers, but as facilitators of shared responsibility and practical alignment.

Building a shared understanding to protect clinical data

I’ve been constantly impressed by the humility and collaboration within this consortium. Our partners brought extraordinary experience — deep knowledge of diseases, research protocols, patient engagement — but never once used it to dominate a conversation. Instead, they offered their insights in ways that strengthened what we were doing together. It created an environment where it felt safe to explore uncertainties and work toward clarity as a group; that’s not something I take for granted.

A lot of our work came down to experience and practicality. We didn’t start from scratch; instead, we used proven tools like data protection impact assessment (DPIA) templates that have been refined over years of working with clinical and health data. We also leaned on existing resources from the European Commission, such as their templates for joint controller agreements.

The goal was never to burden partners with extra processes, but to help legal, regulatory and data protection teams do their work more effectively — having worked in one of those teams myself, I know how overstretched they can be. Offering well-prepared templates, asking who their data protection officers are, and building relationships early on made it easier to move quickly while still doing things properly.

Ultimately, we were able to support a governance model that respected both innovation and institutional due process. We made it clear that while we would conduct a project-wide DPIA, each partner still had to do their own. We weren’t dictating from the top, we were building a toolkit for everyone to adapt and use.

Main takeaways from this 4-year journey

Personally and professionally, GenoMed4All has left a lasting impression on me. It’s deepened my appreciation for due process — not as red tape, but as the structure that enables meaningful innovation. In the rush to develop AI and data-driven tools, it’s easy to dismiss procedure as a barrier. But in truth, those processes are what make responsible, impactful work possible.

This project has also reinforced the importance of humility and collaboration. Though I’ve often said “I lead this” or “I lead that,” the truth is that leadership in GenoMed4All has always been shared. Every success we’ve had has come from people stepping up with their own expertise and contributing to something bigger. It’s a model I’ll carry with me, and one I hope to replicate in future work.

And perhaps most importantly, this project has reminded me that governance itself must evolve. We can’t apply outdated, intrusive models to fast-moving, collaborative, digital health environments; we need innovation in ethics and data protection too. GenoMed4All has shown that it’s possible, and I’m proud to have been a part of this journey.

The legacy of GenoMed4All going forward

As the GenoMed4All project draws to a close, I’ve found myself reflecting on what we’ve built together and what we’re leaving behind. Now, as we finalise our deliverables and recommendations, I hope the tools and insights we’ve developed will serve as a foundation for others navigating this space, particularly as new regulations like the AI Act begin to take shape.

We may not fully understand the impact of that legislation for another few years, once implementing acts are in place and case law starts to form, but the work we’ve done here gives people a chance to begin thinking critically about how AI can and should be used in healthcare — and just as importantly, what its limitations are.

Teresa García Lezana is a Scientific Project Manager at European Genome-phenome Archive, which is part of the Centre for Genomic Regulation (CRG) based in Barcelona. Among her many duties, she provides support for the implementation of fair principles for genomic data management in European projects, especially in healthcare and medical R&I initiatives like GenoMed4All. Her background in translational medicine is supported by a degree in Biology, a PhD focused on liver diseases, and her current postdoctoral research in liver cancer genetics.

As part of our new series of interviews with GenoMed4All partners, Teresa reflects on the role she’s had during the project, her contributions and main learnings, and her expectations for the project’s impact after it’s officially completed in June 2025.

Involvement across work packages and interdisciplinary collaborations

In Genomed4All, the CRG team leads the Work Package 8, which aims to enhance the value of what has been done in the rest of the technical work packages by other project partners.

Our core function revolves around genomic standardization, and the final outcome is the development of the genomic standardization guidelines. For this, we have been interviewing and collecting information from the bioinformaticians involved in the project, to really understand. So we really need what kind of data analysis and formats they’ve been working with.

Through a process of sequencing and standardization, we have created this guide so other peers outside GenoMed4All can understand what has been done in the project, and can follow the same procedures, so eventually they could integrate their data with the data from GenoMed4All. Essentially, this activity allows for our data to be more interoperable and reusable.

Impact on precision medicine for hematology

As part of the CRG team, our main contribution has been towards achieving the fair usage of the data. By working together with other partners in our work package, we’ve created protocols for data access so it can be ethically found and used by other researchers.

Achieving the interoperability of the data – so it can be understood, reused and integrated by others – is a big advancement towards achieving a well-established system for precision medicine in hematology in Europe. Our hope is that the genomic standardization guidelines we’ve created will be a useful resource for the healthcare and medical community in general.

Main takeaways from this 4-year journey

GenoMed4All has made me more aware of how heterogeneous genomic data is, and how complex it is to use and reuse data that has been generated in a clinical context.

When working with different hospitals, different infrastructures and formats, the integration of this data is really complex and there are many challenges that we still need to address; so through this experience I’ve come to realize about the state in which we are in as a community in terms of data standardization.

Also, on a personal level, I’ve learned about the importance of multidisciplinary collaboration. Without the constant cooperation between researchers, clinicians and other key stakeholders, we wouldn’t have been able to achieve what we have done in GenoMed4All.

The legacy of GenoMed4All going forward

One of main legacies of GenoMed4All will be the data that have been created and curated, and that will be accessible to others.

I think the project is also setting some ground for the development of AI and precision medicine that will be used by other projects; all of us together, as a consortium, have created everlasting knowledge that can inspire others and pave the way for future developments in this area.

Catalina González Martín is an International Project Manager at Dedalus. With a background in biomedical engineering, she now manages several projects focused on the design and integration of software solutions to aid the digitalization of healthcare systems worldwide.

As part of our new series of interviews with GenoMed4All partners, Catalina reflects on the role she’s had during the project, her contributions and main learnings, and her expectations for the project’s impact after it’s officially completed in June 2025.

Involvement across work packages and interdisciplinary collaborations

Dedalus has been working mostly across two work packages (WPs): as a leader for WP5 and as a collaborator on WP4.

These two were closely related and required direct collaboration with some of our partners, especially Humanitas Research Hospital and Universidad Politecnica de Madrid (UPM), and of course all the rest of partners that have been providing the data — which has been crucial for the success of both work packages. 

Our main work has been related to the development, the deployment and the validation of the federated learning platform and the homogenization platform. These two platforms serve as the foundation for the implementation, the anonymization and the training of the data models and the algorithms that have been developed in the other WPs that are part of GenoMed4All.

We have also worked very extensively in the treatment of the different data that the hospitals and the other clinical partners have been providing throughout the project, to make sure it could be integrated with the platform while following careful procedures to standardize and anonymize the data.

Impact on precision medicine for hematology

Both work packages have significantly helped the advancement of the federated learning platform, whose ultimate goal is to enhance personalized medicine strategies for hematological diseases. For example, the anonymization of the data is crucial to safeguard the patient’s privacy and for the clinical partners that have provided this data to feed the algorithms. 

Even though we have already seen some positive results before the end of the project, I believe the true impact of our work will likely be perceived over time, especially if the platform continues to be trained with new data, or with new models being implemented.

Main takeaways from this 4-year journey

GenoMed4All has been one of the largest and most heterogeneous projects I’ve been involved in. I have certainly learned a lot, both technically and functionally.

As a Project Manager, for me it’s all about organization and coordination. In such a large project as this one, I have learned that your greatest strength can sometimes also be your greatest weakness if you don’t manage things correctly. So I have realized that when you have many different partners from many different countries, languages, background, different professional profiles, the biggest challenge is to know how to manage all this potential together

The main highlight for me has been taking on this challenge, and I think that we have managed to overcome it and to get the best out of each collaborator.

The legacy of GenoMed4All going forward

Through the collaboration and the connection of all the partners in Europe, in GenoMed4All we have created a very powerful tool, and it’s really important to take care of this legacy. Scientifically and functionally, we’ve brought together so many different expertises, and I believe the involvement of such a diverse group of professionals has enriched the evolution of the project from the very beginning.

Also, thinking about the formation of the algorithms and the joint work we’ve done to create the platform, makes this a tool with great potential, one that can be further explored and tested even after the project has finished. So I definitely would like to see that it continues being implemented in the hospital nodes where it has already been deployed, and they continue training it with different new data models, different algorithms and taking advantage of all its full potential.