AI-Mind trains on 1,000 patients to flag dementia risk early

The AI-Mind consortium pulled brain and cognitive data from 1,000 European citizens across more than 50,000 hours of recording, aiming to spot dementia risk years before a clinical diagnosis. The first phase of the EU-funded project wrapped in early 2026, according to the European Commission's Horizon Magazine.

"By the time cognitive decline has progressed to dementia, opportunities to intervene may already have been lost," says Dr Ira Haraldsen, the neurologist at Oslo University Hospital who coordinates the project. "The point of AI-Mind is to identify the patient early on, before clinical symptoms occur."

The team built two tools. The AI-Mind Connector reads brain electrical activity patterns. The AI-Mind Predictor combines cognitive function, biological markers, medical history, age, education and residence data. "By bringing together all these different aspects, we can identify patterns that help predict a person's risk of developing dementia over the next two to five years," says Vebjørn Andersson, a PhD student and dementia researcher on the team.

Andersson is careful about what the software is for. "It's not supposed to make the diagnosis and then leave it up to the healthcare professionals to treat. It needs to be a human-based, human-centred decision."

Eli Lilly, Roche and Bosch are named as industry partners. Approval as medical devices under EU Medical Device Regulation is expected by 2035.