Periodic Reporting for period 2 - FAIR4Health (Improving Health Research in EU through FAIR Data) Reporting period: 2020-03-01 to 2021-11-30 Summary of the context and overall objectives of the project Due to the nature of health data, its sharing and reuse for research are limited by ethical, legal and technical barriers. The FAIR4Health project facilitated and promoted the application of FAIR principles in health research data, derived from the publicly funded health research initiatives to make them Findable, Accessible, Interoperable, and Reusable (FAIR). To confirm the feasibility of the FAIR4Health solution, we performed two pathfinder case studies to carry out federated machine learning algorithms on FAIRified datasets from five health research organizations. The case studies demonstrated the potential impact of the developed FAIR4Health solution on health outcomes and social care research. Finally, we promoted the FAIRified data to share and reuse in the European Union Health Research community, defining an effective EU-wide strategy for the use of FAIR principles in health research and preparing the ground for a roadmap for health research institutions to offer access to certified FAIR datasets. FAIR4Health has been carried out by 16 partners from 10 different countries over 3 years. FAIR4Health brought together expertise from different domains (health research, data managers, medical informatics, software developers, standards and lawyers). Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far We have been working to share access to high-quality data on health research and healthcare data and its re-use in a trust, controlled and legally compliant environment using Privacy-Preserving Data Mining technologies, which have made feasible the project's goal of accelerating knowledge discovery while reducing bias and improving the strength and quality of the scientific evidence raised by the FAIR4Health community. Furthermore, a community of data scientists from both public research institutions and private companies has been attracted to develop advanced analytical solutions able to communicate with the FAIR4Health platform in order to provide data-driven innovative services that enable a seamlessly application of the new evidence raised into the clinical practice.Finally, the key achieved project outputs that have been already realised with a potential impact are listed below:1. FAIRification Workflow (doi.org/10.1055/s-0040-1713684) to provide a conceptual blueprint for how sensitive, protected patient health information can be made available for medical research in a secure and compliant manner.2. Open-source softwarePublished in the FAIR4Health GitHub repository (github.com/fair4health) under Apache 2.0 licence. • Data Curation and Data Privacy Tools, to technically implement the FAIRification Workflow.• FAIR4Health Platform and Privacy-Preserving Distributed Data Mining service, to support popular algorithms of machine learning for deriving new knowledge, allowing to train statistical/machine learning models using the FAIRified datasets of different health care/research data sources. • FAIR4Health Common Data Model, to include the configuration files for onfhir.io FHIR repository.Published in the bio.tools repository (https://bio.tools/) anchored within ELIXIR, the European Infrastructure for Biological Information, under the GNU General Public Licence v3.0 (GPL-3.0).• Data Curation Tool. https://bio.tools/fair4health_data_curation_tool• Data Privacy Tool. https://bio.tools/fair4health_data_privacy_tool3. FAIR4Health Open Community (fair4health.eu/en/membership) was created to share the results, lessons learned and work done with stakeholders.4. HL7 ‘FAIRness for FHIR’ project (confluence.hl7.org/pages/viewpage.action?pageId=91991234) to develop a new HL7 FHIR Implementation Guide called FHIR4FAIR (build.fhir.org/ig/HL7/fhir-for-fair). 5. RDA WG on Raising FAIRness in health data and HRPOs. The work carried out was expanded to the community through the Research Data Alliance (RDA), creating a working group (WG) on Raising FAIRness in health data and Health Research Performing Organisations (HRPOs), drafting guidelines to implement FAIR data policies in HRPOs (rd-alliance.org/groups/raising-fairness-health-data-and-health-research-performing-organisations-hrpos).6. New European Federation for Medical Informatics (EFMI) WG “FAIR Data” (efmi.org/workinggroups/fair-fair-data-in-health-research-performing-organizations-hrpos).7. Working item proposal for Standardization Technical Committee 251 (CEN TC/251).8. Publication of FAIR4Health metadata in OpenAIRE (explore.openaire.eu/search/other?pid=10.5281%2Fzenodo.6401048) Digital CSIC (doi.org/10.20350/digitalCSIC/14571) Zenodo (doi.org/10.5281/zenodo.6401048) and GitHub (github.com/fair4health/metadata).9. Scientific publications at JCR journals. Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far) The addressed societal problem is to facilitate the sharing and reuse of health data to facilitate scientific discovery, as well as the advancement of outcomes. The societal implications are many-fold and can have consequences for health emergencies that could develop rapidly and extensively across borders. One example would be the Covid crisis [doi.org/10.1371/journal.ppat.1008902 doi.org/10.1186/s12874-021-01304-y]. There are several other diseases that could be targeted for concerted action using this model and the volumes of data that each produces would need to be managed carefully. FAIR4Health encourages the scientific community to apply the FAIR principles to that purpose, sharing knowledge and results from previous research and contributing to open science. There is a large imbalance between developed nations and low- and middle-income (LMIC) countries in terms of available healthcare, this partly contributes to the disparities that can be seen in the levels of health research, and which is also a factor of lack of funding and infrastructure. In today’s digital world, LMICs could also benefit from shared resources, this can be achieved through the application of the same FAIR principles being widely adopted in developed nations. The report published by the European Commission on Cost-benefit analysis for FAIR research data and the Cost of not having FAIR research data (data.europa.eu/doi/10.2777/02999) highlights the importance of applying the FAIR principles in health in the economic and social field.To achieve appropriate exploitation results, the project received support from the Horizon Results Booster program (horizonresultsbooster.eu) and great results were achieved in terms of Portfolio Dissemination & Exploitation Strategy, and Business Plan Development.FAIR4Health had four specific objectives, all successfully achieved at the end of the project:1. Effective dissemination strategy at the EU level, which will keep active after the project end, since we will keep working on FAIR principles implementation.2. FAIR data certification system with essential collaboration with RDA through HL7: we promoted the FAIRified data to share and reuse in the European Union Health Research community, defining an effective EU-wide strategy for the use of FAIR principles in health research and preparing the ground for a roadmap for health research institutions to offer access to certified FAIR datasets.3. Intuitive FAIR4Health platform under development with proof of concept: the developed platform allowed us to implement federated machine learning algorithms for the performance of the FAIR4Health pathfinder case studies.4. Clinical validation of two pathfinder case studies.Address (URL) of the project's public websitehttps://www.fair4health.eu/ whatsapp-image-2021-11-24-at-10-37-03.jpeg FAIR4Health Logo
