Cross-border collaboration can tackle the challenges in accessing relevant health data essential for international collaboration between scientists and clinicians, researchers, and health industry. Privacy concerns and regulations on personal data have made the sharing of health data increasingly complex and time-consuming for data controllers, thus severely limiting the access of SMEs, researchers, and innovators to health data. Further complications in cross-border collaboration arise from differences in interpreting the EU GDPR, national regulations, and heterogenous and changing data permit processes at hospital sites. The PHEMS project will provide European children’s hospitals with a decentralized and open health data ecosystem concept consisting of technical components and governance frameworks. The objective is to facilitate access to health data, advance federated health data analysis and build services for the on-demand generation of shareable, synthetized, and anonymized datasets. To achieve this, the project will focus on bridging the gaps in data access and use, especially in the integration of ethical, legal, and technical requirements, including the responsibilities of data controllers and the rights of data subjects. This will allow health data controllers to engage in collaboration without losing control on compliance with respect to GDPR, national legislation or internal policies of their organization. The techniques and tools for generating algorithmically anonymized and synthetic datasets will undergo robust validation processes through three clinical use cases conducted by the European Children’s Hospitals Organisation (ECHO) community. The goal is to assess the usage of custom-generated synthetic data with real-life questions. Data users, such as researchers, SMEs, innovators and the pharmaceutical and MedTech industry, will be engaged through community building, hackathons, and interaction with relevant European large-scale initiatives.

Enabling integration of medical and research data, secure data sharing and leveraging responsible state-of-the-art artificial intelligence (AI)-mediated models opens immense possibilities to mitigate the impact of chronic immune-mediated diseases (CIMDs) affecting 10% of Europeans. Eight European universities, leaders in the medical and analytical field, three SMEs, one research institute and one company, at the forefront of clinical AI implementation, data infrastructure, and security, and a Patient Organisation formed the consortium WISDOM. The consortium's overarching aim is to convert complex biological information from the existing data sources into actionable insights. WISDOM builds on the premise that computational tools can provide valuable knowledge and guide decision-making at critical stages in the individual patient journey, from diagnosis to treatment initiation and optimisation. To unlock the potential of the existing data, WISDOM will address barriers of data integration and accessibility and deploy novel approaches for data processing, harmonisation, integration, and secure, trustworthy data sharing with federated access. WISDOM aims to develop computational risk stratification and outcome prediction models and tools in different CIMD use cases, building on large EU-funded multimodal datasets, and prospectively validate them on technical, clinical and user aspects to facilitate data-driven and patient-focused diagnosis, treatment, and monitoring. WISDOM aims to promote the widespread utilisation of data and facilitate responsible and critical assessment of the use of AI in healthcare using an end-user guided approach leveraging collaboration among clinicians, researchers, legal and AI experts, patient associations and rich stakeholder expertise. WISDOM’s ultimate ambition is to revolutionise the integration, management, and analysis of health data across diseases and borders to promote personalised interventions and well-being.

The goals of this inter-disciplinary project are to 1) gain understanding of the mechanisms causing chemoresistance in high-grade serous ovarian cancer (HGSOC) patients, 2) deliver tools that enable effective and cost-efficient personalised treatment options for HGSOC patients, and 3) commercialise predictive kits & software for treatment response prediction and finding the right therapeutic regimen to the right patient. This project takes an advantage on prospectively and longitudinally collected fresh and blood specimens of HGSOC patients. Longitudinal, multi-layer data are analysed with ML and AI methods to predict patient treatment response and identify the most effective treatment options. Drug screening with patient-derived 3D ex vivo cell models are used to identify drug combination options. Key results will be validated with retrospective cohorts, and in vitro, ex vivo & in vivo models. We will develop an open-source software to visualise all relevant patient-specific data to guide clinical decision-making. Clinically most actionable treatment suggestions will be evaluated in virtual molecular tumour board and translated to patient care. Ovarian cancer kills more than 44,000 women in Europe every year due to lack of effective and long-lasting therapeutic regimens. DECIDER presents an innovative strategy to suggest effective treatments that lead to a marked decrease in ovarian cancer deaths and reduce the number of expensive but inefficient treatments. Our approach paves the way to move beyond the current trial-and-error clinical assessment of drug combinations toward more systematic prediction of the most effective treatments for each patient. The proposed concept will be a major breakthrough in personalised medicine and will benefit individual patients and the health-care system through more effective treatments, and the diagnostic and pharmaceutical industry through tools for better stratified clinical trials, and novel treatment and diagnostic modalities.