{"id":34147,"date":"2025-10-30T11:25:49","date_gmt":"2025-10-30T09:25:49","guid":{"rendered":"https:\/\/ticsalutsocial.atoom.space\/noticia\/search-avanca-en-la-creacio-de-dades-sintetiques-per-impulsar-la-recerca-en-salut-a-europa\/"},"modified":"2025-11-03T11:27:54","modified_gmt":"2025-11-03T09:27:54","slug":"search-avanca-en-la-creacio-de-dades-sintetiques-per-impulsar-la-recerca-en-salut-a-europa","status":"publish","type":"noticia","link":"https:\/\/ticsalutsocial.atoom.space\/en\/noticia\/search-avanca-en-la-creacio-de-dades-sintetiques-per-impulsar-la-recerca-en-salut-a-europa\/","title":{"rendered":"SEARCH advances in the creation of synthetic data to drive health research in Europe"},"content":{"rendered":"\n

More than 50 representatives from the 28 institutions forming the SEARCH<\/a> consortium, including the TIC Salut i Social Foundation, attended the plenary meeting held in Athens, Greece, on 23\u201324 October to mark the beginning of the project\u2019s second year. The meeting provided an opportunity to review the main milestones achieved so far and to discuss the roadmap for the coming 12 months.<\/p>\n\n\n\n

The working sessions addressed key topics related to the generation of synthetic medical datasets and their role in supporting clinical decision-making, as well as collaboration with industry and the effective communication and dissemination of results. Discussions also focused on areas such as data mapping and ongoing consultation, the creation of a federated platform for FAIR (Findable, Accessible, Interoperable and Reusable) data discovery, and distributed machine learning, among others.<\/p>\n\n\n\n

Achievements and next steps<\/h2>\n\n\n\n

The SEARCH project has successfully reached the main milestones planned for its first year, consolidating the technical, clinical, ethical, and governance foundations required for the development of its synthetic health data platform<\/strong>.<\/p>\n\n\n\n

Key achievements include the definition of technical and user requirements through stakeholder consultation; the creation of a dataset registry of more than 140 datasets with metadata and usability mapping for federation and synthetic generation; the design of the platform\u2019s architecture and interoperability specifications; the release of the first version of the Synthetic Data Assessment Framework; and the implementation of initial federated artificial intelligence methodologies for multimodal data generation.<\/p>\n\n\n\n

Regarding clinical pilots, the initial study initiation packages have been prepared for validation studies in the project\u2019s therapeutic areas: cardiovascular, stroke, gastrointestinal, and gynaecological oncology.<\/p>\n\n\n\n

Overall, during its first year, the SEARCH project has advanced consistently with its overarching mission: to create an integrated, secure, and privacy-preserving ecosystem that enables the sharing and use of health data across sectors<\/strong>, unlocking new possibilities in diagnostics, treatment, and personalised medicine. Although major progress has been made, the project still has three years ahead to fully achieve its vision.<\/p>\n\n\n\n

At the Athens meeting, participants identified the main challenges for the next 12 months, including the implementation of the federated architecture, enhancing interoperability and generating trustworthy synthetic data, obtaining ethical and regulatory approvals, adapting to new European regulations, and strengthening the project\u2019s long-term sustainability and impact.<\/p>\n\n\n\n

The role of the TIC Salut i Social<\/strong> Foundation<\/h2>\n\n\n\n

The TIC Salut i Social Foundation plays an active role in the SEARCH project, contributing to the design of evaluation frameworks to ensure the credibility and safety of synthetic data used in AI model development.<\/p>\n\n\n\n

As explained by Vaneza G\u00f3mez<\/strong>, member of the Foundation\u2019s Artificial Intelligence team, in a video recorded at the SEARCH annual meeting, \u201cduring this first year, we have contributed to research on explainability methods for synthetic data generation models, ensuring that all synthetic data produced within the project can be explainable, useful, and reliable for all stakeholders involved.\u201d<\/strong><\/p>\n\n\n\n

\n