HDR UK Gateway
HDR Gateway logo

Bookmarks

Evaluating Brain Health across the life course: exploring SDE capabilities for machine learning-driven, imaging-based, dementia diagnostics

Safe People

Organisation name

University Hospital Southampton NHS Foundation Trust

Applicant name(s)

Funders/ Sponsors

Safe Projects

Project ID

SDE_W_PROJ_87

Lay summary

Perfusion SPECT imaging is a type of brain scan that shows how blood flows through different areas of the brain. In conditions such as Alzheimer’s disease, this blood flow can change. SPECT scans can sometimes detect these changes before any visible damage appears on other types of scans. This makes them a useful early warning tool that could help doctors start treatment sooner, when it is more likely to be effective. In Alzheimer’s disease, lower blood flow in areas such as the parietal and temporal lobes is an important sign of the condition. Detecting this early provides doctors with vital information about how the disease is developing and supports more confident diagnosis. By combining SPECT results with information from other scans and medical records, researchers can better understand how the disease affects different people. This could help doctors group patients more accurately, plan more effective treatments, and diagnose Alzheimer’s earlier. We also plan to look at people’s wider health information collected over their lifetime. This could help us understand which factors influence the risk of Alzheimer’s, how people respond to treatment, and what their likely outcomes might be. Our research will use imaging data from the SWASH+ PACS consortium, which includes Salisbury, Southampton, Portsmouth, and the Isle of Wight. In the past year, Hampshire Hospitals (HHFT) has joined the consortium, meaning the work now covers the whole Hampshire and Isle of Wight Integrated Care Board (HIOW ICB) area. Many patients may come to University Hospital Southampton (UHS) for specialist neurological care but will have had other tests at the partner hospitals. By safely linking this information, we aim to improve how Alzheimer’s disease is detected and predicted. We will begin by working with colleagues at HHFT and hope to expand this approach across other sites in the future.

Public benefit statement

New advances in machine learning could help doctors diagnose conditions like Alzheimer’s disease earlier and with less invasive tests. At the moment, some diagnostic tests require samples of cerebrospinal fluid (CSF), which involves a lumbar puncture. This research will explore whether brain scan images can provide similar information, reducing the need for such procedures in the future. Using the Secure Data Environment (SDE), the project will safely link different types of brain imaging data collected over people’s lives with existing CSF results. By studying these links, researchers hope to understand how changes seen in scans relate to biological markers of disease. The findings will help build the foundation for new, non-invasive ways to detect Alzheimer’s and related conditions earlier, and to develop more personalised approaches to treatment and care.

Other approval committees

Latest approval date

04/09/2025

Safe Data

Dataset(s) name

CSF Biomarker analysis; MRI and SPECT Imaging

Safe Setting

Access type

TRE

Safe Outputs

Link to research outputs