This AI tool may outperform clinical tests at detecting Alzheimer's

New Delhi, July 13
Scientists at the Cambridge University in the UK have developed an artificial intelligence (AI)-based tool that can help detect patients with early dementia whether they will remain stable or develop Alzheimer's.

Dementia is a global healthcare challenge, affecting over 55 million people at an estimated annual cost of $820 billion.

The cases are expected to almost triple over the next 50 years.

To develop the new AI model, the researchers used routinely gathered, non-invasive and low-cost patient data - cognitive tests and structural MRI scans revealing grey matter shrinkage - from over 400 participants in a research cohort in the US.

They next evaluated the model with real-world patient data from an additional 600 participants in the US cohort, as well as longitudinal data from 900 persons in memory clinics in the United Kingdom and Singapore.

The algorithm was able to identify between persons with stable mild cognitive impairment and those who developed Alzheimer's disease within three years, according to the story published in the journal EClinicalMedicine.

It successfully identified those who developed Alzheimer's in 82 per cent of cases and those who did not in 81 per cent of cases using only cognitive tests and an MRI scan which provides hope that this model may well be accurate.

Professor Zoe Kourtzi from the Department of Psychology at the University of Cambridge said this tool would be good at predicting whether or not someone will progress to Alzheimer’s, and that since it has been tested in real life too, generalisations can be made.

Ben Underwood from the University of Cambridge said this would help alleviate many existing concerns for the patients and their families.