Artificial intelligence unlocks secret to Alzheimer’s identity

A test to identify which form of Alzheimer’s Disease a patient is suffering from could be on the NHS within three years, medics believe after a breakthrough in analysing brain scans.

Tests for Alzheimer’s types can until now only be carried out after death, but the new technique offers the prospect of identifying not just the precise form of the disease in live patients, but also what stage.

Researchers developed an artificial intelligence (AI) algorithm to analyse brain scans and found it could identify the disease in patients with unprecedented accuracy and detail. The breakthrough offers the prospect of the disease being targeted much more effectively. Medics could more easily understand which drugs have the best chance of working if they know the variety of Alzheimer’s they are tackling. It might also be possible to use the AI tool to help provide diagnosis of the disease in its early stages.

Medics also believe the AI technique can be used in tackling a several other diseases, including chronic obstructive pulmonary disease (COPD) and multiple sclerosis (MS), which like dementia are characterised by a long and slow series of changes. The AI tool analyses brain scans with an algorithm that has been developed by using new machine learning techniques to process thousands of MRI scans of living Alzheimer’s patients.

Professor Daniel Alexander said: “This new algorithm has the unique ability to reveal groups of patients with different variants of disease.”