The VISTA project is helping to bring the power of artificial intelligence (AI) and machine learning safely into health data research – enabling larger, more complex studies while keeping sensitive data secure.
As research grows in scale and sophistication, Trusted Research Environments (TREs) – also known as Secure Data Environments (SDEs) – must evolve to support new methods and technologies. This includes making sure that advanced tools and technologies like Artificial Intelligence (AI) can be used responsibly, and that outputs from data research are carefully checked before leaving the secure environment to ensure sensitive data is not inadvertently released.
VISTA (Viable Implementation of SATRE, SACRO and Tools for AI) is one of the DARE UK Early Adopter projects launched in March 2025. Led by Cambridge University Hospitals NHS Foundation Trust in partnership with Health Innovation East and Cambridge University Health Partners, VISTA is enhancing the East of England Secure Data Environment (EoE-SDE) by piloting key TREvolution standards and AI capabilities.
At the heart of the project are three elements:
- SATRE (Standardised Architecture for Trusted Research Environments) – a framework for building TREs to a consistent, high-quality standard.
- SACRO (Semi-Automated Checking of Research Outputs) – tools that help identify potential privacy risks in research outputs before results are released.
- AI and machine learning – methods that can unlock new insights but need careful governance to ensure they are used safely and responsibly.
What VISTA hopes to achieve
By the end of the project, the VISTA team aims to show how EoE-SDE can support large-scale, AI-enabled research safely, efficiently, and consistently, while also helping other TREs adopt these approaches. The team aims to:
- Benchmark SATRE – delivering a gold-standard assessment of how the East of England SDE aligns with this reference architecture, helping guide its further development.
- Advance semi-automated output checking – testing and refining available research output checking tools in real research settings, making the process of sharing results faster, safer, and more consistent across TREs.
- Enable safe AI training – developing policies and tools that can allow AI and machine learning models to be trained on sensitive data within TREs, and safely exported for use in real-world health research.
- Prepare for NHS adoption – supporting the readiness of AI models for use in NHS services, paving the way for new treatments, insights, and patient benefits.
Progress so far
In its first few months, VISTA has already achieved significant milestones:
- Achieved full marks on 88% of mandatory clauses in its first SATRE assessment, and at least partial marks on all areas.
- Begun co-leading a national working group to align SATRE with other standards such as ISO 27001 (an internationally recognised standard for information security management systems) and the Data Security and Protection Toolkit (DSPT) (an online self-assessment tool for legal and regulatory compliance in handling personal data), supporting easier compliance across TREs.
- Identified real-life research test cases in heart failure and cardiometabolic disease.
- Installed and trialled AI disclosure checking tools within the East of England SDE
- Created training materials to help researchers use semi-automated research output checking tools effectively.
- Advanced safe AI research by piloting ‘SACRO-ML’ (a set of tools that helps make sure machine learning models trained on sensitive data are safe to share publicly)with sample models and developing a risk assessment policy for responsible AI training.
- Actively engaged the public and stakeholders through forums, webinars, and advisory groups to ensure public perspectives shape the work.
How VISTA supports TREvolution
VISTA is central to TREvolution, the DARE UK project transforming how TREs operate across the country. By testing and refining standards like SATRE, piloting tools like SACRO, and creating policies for safe AI use, the project is helping to ensure TREs are equipped to handle the next generation of research.
Together with the other Early Adopters, VISTA is showing how secure data environments can adapt to new technologies – making health data research faster, safer, and more innovative, while keeping public trust at the core.
About the DARE UK Early Adopters
The DARE UK Early Adopters are pioneering projects helping to put the standards and innovations assembled through TREvolution into practice. Working across more than 10 Trusted Research Environments (TREs), including five sub-national NHS SDEs in England, these projects bring together researchers, TRE operators and members of the public. Their role is to test the new approaches in real-world settings, making sure they are practical, effective and safe before being more widely adopted for research in the public interest.
