NOVEL research has demonstrated that machine learning technology can be used to identify the most accurate means of predicting advancement in Alzheimer’s disease. The model demonstrated the value of MRI as a powerful prognostic tool for both cognitively normal, asymptomatic individuals, and those with mild cognitive impairment. It further concluded that radiological tools for tracking molecular biomarkers of Alzheimer’s disease, such as PET scans, were more useful in individuals already experiencing mild cognitive impairment.
Progression in Alzheimer’s can be variable, with symptom onset sometimes occurring over decades, whereas other individuals decline rapidly, creating challenges for predicting the rate of disease advancement.
“When we can confidently say someone has dementia, it is too late. A lot of damage has already happened to the brain, and it’s irreversible damage,” stated Mert Sabunca, Radiology, Weill Cornell Medical College, New York City, USA, senior author of the study.
The researchers used neural-network machine learning to analyse 5 years’ worth of data on patients who were either cognitively normal or who had mild cognitive impairment. Data was derived from the Alzheimer’s Disease Neuroimaging Initiative, and encompassed patient data including genetic history, PET, and MRI scans.
After applying the machine learning analysis, the researcher observed several significant correlations, notably, concerning time horizons for disease progression. They found that predicting whether an individual will change from asymptomatic to displaying mild cognitive impairment was much easier over a shorter period of 1 year, compared to 5 years. However, conversely, they found that predicting whether an individual will decline from mild cognitive impairment to Alzheimer’s dementia is most accurate over a timeline of around 4 years.
The data further showed that MRI scans are the most informative predictor for asymptomatic cases and are particularly helpful for estimating whether an individual will develop symptoms over the next 3 years. However, once patients have developed mild cognitive impairment, PET scans appear to be more useful to predict progression.