Case study: PICTURES
InterdisciPlInary Collaboration for efficienT and effective Use of clinical images in big data healthcare RESearch (PICTURES)
Overview
Clinical images such as MRI, CT and X-ray scans hold enormous potential for improving early diagnosis, understanding disease progression, and enabling personalised medicine. In Scotland alone, millions of clinical images are generated each year through routine care, yet historically these “real world” images have been difficult to access for research due to technical, governance and confidentiality challenges.
In addition, most image-based research now uses images collected specifically for particular research projects, which differ from “real world” situations.
Using foundation blocks from previous research grants, the PICTURES project set out to unlock the value of Scotland’s national imaging data by developing secure, scalable infrastructure that enables researchers to safely access and analyse large volumes of clinical images linked to health records. By strengthening Safe Haven capabilities and supporting exemplar studies in dementia and lung cancer, PICTURES demonstrates how population-scale imaging data can be used responsibly to advance AI-driven healthcare research.
Who is involved?
This five-year project was led by the University of Dundee, and funded by EPSRC (Engineering and Physical Sciences Research Council), and MRC (Medical Research Council). Its partners include the University of Edinburgh, University of Abertay, Public Health Scotland, and Aidence B.V.
The multidisciplinary team included clinicians, engineers, imaging specialists, data scientists and industry partners. Secure data access was supported through eDRIS and the Health Informatics Centre.
The challenge
Extended lifespans mean more years living with multiple chronic conditions (multimorbidity), resulting in diminished quality of life with potential life expectancy implications. A major challenge for healthcare and social services is determining optimal support for individuals with multiple health issues.
This challenge is exemplified in determining appropriate treatment for patients with multiple conditions presenting to busy Emergency Departments. Another complication is polypharmacy – long-term use of various medications – which can create additional complications.
What data is being used?
PICTURES focused on enabling access to Scotland’s national clinical imaging database, which contains around 23 million images collected since 2010. These routine scans were linked with other health data to support large-scale research.
What have they found, and what are the next steps?
PICTURES successfully delivered protected infrastructure for population-scale imaging research and supported two exemplar studies: predicting dementia from brain scans and assessing lung nodule malignancy from chest images.
Key achievements include:
- Creating secure pathways for accessing large volumes of clinical imaging data
- Expanding eDRIS cohort-building capabilities
- Developing scalable technology for big data within Safe Haven environments
- Negotiating complex collaboration agreements balancing public and commercial interests
- Establishing a discount licensing model to deliver value back to NHS Scotland
Alongside technical progress, the project highlighted challenges around commercial engagement, regulatory constraints and evolving industry requirements. These experiences have informed future approaches to governance, anonymisation and collaboration with medical technology partners.
The infrastructure and lessons from PICTURES now provide a foundation for future imaging and AI projects across Scotland.
How could this research help inform policies and improve lives?
By enabling secure access to real-world clinical images at national scale, PICTURES supports earlier diagnosis, more accurate risk assessment, and the development of AI tools that reflect routine patient care.
The project provides a model for how imaging data can be used responsibly to advance precision medicine, while protecting patient privacy.