A new UK‑developed artificial intelligence system is being positioned as a potential breakthrough in the early identification of Alzheimer’s disease, aiming to reduce the number of people who remain undiagnosed until symptoms have significantly advanced.
DementAI, created by consultancy Katalyze Data as part of the SAS Hackathon 2025, is designed to detect risk signals already present within existing clinical records. According to the team, retrospective analysis suggests the tool could help identify Alzheimer’s up to around two years earlier than current routes, potentially allowing patients to access assessment and support while interventions remain more effective.
Dementia is the UK’s leading cause of death and many people estimated to be living with the condition still lack a formal diagnosis. This creates what the developers describe as a hidden backlog, where warning signs exist in plain sight but have not yet triggered specialist referral.
Built as an end‑to‑end working prototype, DementAI connects stages that clinicians typically manage separately, from analysing medical record data to deploying models within decision pathways. It is designed to work with information health providers already hold, turning fragmented data into actionable insight without adding new screening burdens.
The system combines structured medical records, brain activity data and unstructured clinical information, using synthetic data where appropriate to support development. By blending these signals, it aims to detect subtle patterns of decline that may be difficult to identify during short consultations.
Recognised as the Health Care & Life Sciences category winner at the SAS Hackathon 2025, the tool embeds analytics within a governed workflow, providing clinicians with a transparent rationale for why a patient has been flagged.
Tamás Bosznay, principal consultant at Katalyze Data, said: “We are in a race against time when it comes to dementia. Early identification can make a meaningful difference to how patients and families experience the condition. But without better ways of finding people sooner, those opportunities can be lost.
“We didn’t build DementAI just to make predictions; we built it to buy patients time. By surfacing the signals already hiding in plain sight within clinical records, the system is designed to help ensure that when care teams are ready to act, the right patients are identified earlier and more consistently.”
Dr Iain Brown, global head of AI & data science at SAS, explained: “Synthetic data, agentic AI concepts and governance are not ‘nice‑to‑haves’ in sensitive settings like healthcare – they are what make innovation usable at scale. DementAI shows how artificial intelligence can be applied in a way that is both ambitious and responsible.”
The team is now seeking engagement with NHS Trusts to explore pilot deployments that could validate the model’s impact and support efforts to reduce diagnostic delays.
