Early Adopters
VISTA: Viable Implementation of SATRE, SACRO and Tools for AI
Scaling research using sensitive data in Trusted Research Environments (TREs) is complex, with challenges in automation, disclosure control, and AI model training. VISTA supported efforts to tackle these challenges by piloting key TREvolution standards and capabilities.
VISTA supported the East of England Secure Data Environment (SDE) (one of 11 sub-national SDEs in England) to enhance its existing processes, by accelerating the adoption of Al and machine learning capabilities and improving its capacity for large-scale research projects involving sensitive data. This will enable complex research at scale, increasing efficiency and driving new scientific discoveries to benefit patients.
Drawing on insights from the East of England SDE’s platform development, which follows the SATRE (Standardised Architecture for Trusted Research Environments) reference architecture developed as part of the DARE UK Phase 1 Driver Projects, VISTA helped refine SATRE to ensure it meets the needs of future TRE implementations. The project also tested and assessed Semi-Automated Checking of Research Outputs (SACRO) tooling at scale, to support its readiness for wider adoption.
VISTA’s efforts ultimately enhanced TRE/SDE standards and capabilities, unlocking the possibility of researchers to safely train and export AI and machine learning models, enabling and speeding up new projects and advancing research for patient and public benefit.
By the end of the project, VISTA:
- Delivered a gold-standard benchmarking of a SATRE TRE implementation, supporting its next stage of development.
- Tested and refined semi-automated research output checking within the East of England SDE, improving its readiness for adoption across other TREs.
- Safely trained and exported outputs from AI and machine learning models, enabling previously unfeasible projects to move forward.
- Supported AI model readiness for real-world deployment in NHS services, benefiting patients and the public.
Project information
Lead organisation: Cambridge University Hospitals NHS Foundation Trust
Principal investigator: Dr Alexis Joannides, Chief Research Informatics Officer, Honorary Consultant Neurosurgeon, Co-Director of NIHR HealthTech Research Centre in Brain Injury, Cambridge University Hospitals
Project duration: 14 months
Project partners: Health Innovation East, Cambridge University Health Partners
Funding provided: £393,725
Primary contact email: eoe-sde@healthinnovationeast.co.uk
Relevant links
Project outputs and reports
Final report
Eastern England SDE Statistical Disclosure Control Training Presentation
Eastern England SDE Airlock Manager Statistical Disclosure Control SOP
Eastern England SDE SATRE Self-Assessment: Findings, Insights and Iterations
Eastern England SDE Airlock Manager Model Disclosure Control SOP
Eastern England SDE Researcher Model Disclosure Control SOP
Eastern England SDE Researcher Statistical Disclosure Control SOP
VISTA AI Risk Assessment Toolkit
Eastern England SDE Scaling Safe Outputs with SACRO Integration
Eastern England SDE Safe Models, Safe AI: Governance and Disclosure Testing
Eastern England SDE Public and Patient Involvement and Engagement (PPIE) Learnings
GET IN TOUCH
If you’re interested in learning more about our work, how it can benefit you, or how to get involved, click the button to get in touch with us using our contact form.