Data-driven medicine is fundamental to improving the accessibility and quality of the healthcare system. The availability of data is crucial for this purpose. In the context of a distributed analytics platform for analyzing healthcare data - employing the Personal Health Train paradigm - we propose to implement a solid data FAIRification infrastructure. This will allow us to achieve findability, accessibility, interoperability, and reusability of data, metadata, and results within a network of several medical centers participating in the BETTER Horizon Europe project, where the study of rare diseases (such as intellectual disability and inherited retinal dystrophies) will be targeted. Impacts will be visible to a large population of healthcare practitioners, prospectively influencing health policymakers.
Analytics paradigm called the Personal Health Train (PHT) [
As a technical partner of the project, Politecnico di Milano (i.e., the authors) will particularly
focus on BETTER’s objective to guide medical centers in collecting patients’ data following
a common schema in order to promote interoperability and re-use of datasets in scope. This
includes legal/ethical data protection authorizations as well as data FAIRification (data
documentation, cataloging, and mapping to well-established ontologies [
Importantly, the project will aim at the integration of external sources such as European Health Data Space (EHDS), the 1+Million Genomes initiative (1+MG), and the European Open Science Cloud. Legal and ethical implications will be duly considered, and data access and re-use procedures will be proposed. Data pseudonymization will be performed as a default preprocessing step, mitigating the risk of personal data leaks. A real-world large-scale data integration framework (based on well-established ontologies) will be demonstrated taking into account heterogeneous datasets. The platform will be tested primarily on two rare disease use cases, inherent to pediatric intellectual disability and inherited retinal dystrophies.
In conclusion, the BETTER project relies on “bringing computation to data” via incremental and federated learning, which avoids unnecessary data moving across medical centers while exploiting much of the information encoded in such data. The project will enable EU medical centers and beyond to make full use of the potential ofered by a safe and secure exchange, use, and reuse of health data fostered by robust data FAIRification. In the context of intellectual disability and inherited retinal dystrophies – with the potential of expanding the same paradigm to other diseases – the generated analytical tools will help healthcare professionals become more proficient in cutting-edge digital technologies, data-driven decision support, health risk surveillance, and control activities, monitoring and management of healthcare quality levels. Acknowledgements. The work is supported by BETTER, Grant agreement 101136262.