Sophia Antipolis, 25 March, 2025: The DataTools4Heart (DT4H) project, which focuses on enhancing the lifestyle of cardiovascular disease patients by creating a secure, privacy-preserving toolkit for data reuse in cardiology, has successfully completed its development phase and will aim to reach key milestones in 2025. The project aims to drive innovation in healthcare technology and foster collaborations that bridge clinical and computational expertise.
Cardiovascular disease is the leading cause of death globally. Until now, cardiology data that could advance research and healthcare has remained unused in hospitals across Europe. This is due to data privacy requirements and variations in data formats and languages. To tackle these challenges, DT4H will extract, translate, and reuse data in a federated manner, i.e. without the information being shared with anyone or transferred out of the hospital.
“The vision for the toolbox is to provide an all-in-one, user-friendly platform where clinicians and researchers can easily access structured clinical data and perform AI-powered analyses—no programming skills will be required. The initial version is set to launch by the end of 2025, marking the beginning of an iterative process to refine the toolbox and adapt it to meet the specific needs of clinicians," says Cristian Izquierdo, scientific lead of the project at the University of Barcelona.
“Funded by the European Union’s Horizon Europe Framework, DataTools4Heart is dedicated to creating an all-encompassing cardiology data toolbox designed for clinicians, researchers, and data scientists,” says Laura G. Antiga, a research engineer based at the Barcelona Supercomputing Center, Barcelona, Spain - part of the teams assembling the tools. “The toolbox will offer advanced features, including data ingestion and harmonisation, multilingual natural language processing, federated machine learning, and data synthesis.”
The overarching aim for DT4H is a commitment to advancing federated learning solutions for healthcare. Thus, the consortium is preparing for important developments through the following key actions for 2025:
-
Deployment of Stable Software Versions: By deploying the first stable versions of its software, the project will efficiently manage the federated learning network. This technical milestone represents a significant step forward in enabling secure, privacy-preserving collaboration across clinical and research institutions. The first stable version of the platform is expected to be released by the end of 2025.
-
Integration of Applications into a Unified Interface: To enhance usability and streamline workflows, the project has initiated two dedicated working groups focused on integrating diverse applications into a single, cohesive user interface. This effort ensures that users can access all functionalities seamlessly within one platform. Clinicians and researchers will gain access to powerful functionalities, including the ability to extract meaningful statistics from diverse demographic groups, convert unstructured data into structured formats, and perform AI-driven analyses—all without requiring programming expertise. Additionally, they can retrieve explainable insights that can seamlessly integrate into clinical practice, enhancing decision-making and patient care.
-
Clinician Engagement for Tool Optimization: Usability by physicians, researchers and data scientists is key to the success of DT4H. Recognising the importance of end-user needs, the project has launched a validation and request working group that collaborates closely with clinicians. These sessions aim to gather insights into analytical requirements and usability preferences, ensuring the resulting tools are both practical and user-friendly for clinical environments.
The first clinical analysis using the DT4H platform is anticipated to commence by mid-2025. Eight hospitals covering seven languages (Spanish, Dutch, Swedish, English, Czech, Italian and Romanian) will continue testing the system using actual clinical questions from their outpatient and emergency departments. Multilingual, AI-powered virtual assistants will help clinician researchers to navigate through the platform and data. These eight hospitals are:
The results of these analyses, along with demonstrations of the software, will be unveiled at public forums and conferences later in the year, including the ESC Congress in Madrid (29 August- 1 September). There will be a dedicated session to the project, where the first results with real-world data are expected to be presented to delegates. These events will provide an opportunity to showcase the project's potential to revolutionise healthcare data analysis through federated learning.
Cristian Izquierdo explains: "Federated learning is transforming healthcare by enabling collaboration across institutions without compromising privacy. It builds better AI models from diverse data, improving decision-making and outcomes, while preserving ethical standards." ENDS