Researchers at the University of Cambridge, UK, have developed a new AI tool that predicts Alzheimer's disease progression with 82% accuracy from routinely collected cognitive tests and MRI scans. The corresponding study was published in eClinicalMedicine.
Early diagnosis of Alzheimer's disease and other dementias is linked to more effective treatment and can allow patients and their families to plan for the future. Until now, however, there has been a lack of sensitive tools for stratifying patients early on, meaning that many end up being undiagnosed or misdiagnosed.
In the current study, researchers built an algorithm for distinguishing Alzheimer's disease progression using cognitive tests and structural MRI scans showing gray matter atrophy from over 400 individuals in a US-based research corhort. They then tested the model on data from 600 other participants from the US cohort, as well as longitudinal data from 900 people from memory clinics in the UK and Signapore.
The model was able to distinguish between participants with stable mild cognitive impairment and those who had progressed to Alzheimer's disease over three years. It correctly identified patients who developed Alzheimer's disease in 82% of cases and correctly identified those who didn't 81% of the time.
The algorithm was approximately three times more accurate at predicting progression to Alzheimer's than the current standard of care: standard clinical markers like gray matter atrophy or cognitive scores, or clinical diagnosis.
"AI models are only as good as the data they are trained on. To make sure ours has the potential to be adopted in a health care setting, we trained and tested it on routinely-collected data not just from research cohorts, but from patients in actual memory clinics. This shows it will be generalizable to a real-world setting," said senior study author, Professor Zoe Kourtzi from the Department of Psychology at the University of Cambridge, in a press release.
The researchers now hope to adapt their model for use in other forms of dementia including vascular dementia and frontotemporal dementia, using different types of data such as markers from blood tests. They hope their tool will help clinicians assign patients to the right diagnostic or treatment pathways, including clinical trials to accelerate the discovery of new therapeutics.
Sources: Neuroscience News, eClinicalMedicine
