AI Blood Test for Alzheimer's

As the average human lifespan continues to increase, neurodegenerative diseases like Alzheimer’s disease (AD) are becoming more widespread. And despite an intense research focus, a cure is still elusive, possibly due to the fact that most of the patients who are participating in the trials are already in an advanced stage of the disease.

Now researchers are using machine learning to detect AD earlier, while there is a better likelihood of slowing its progression. In one recent study, a peptide known as amyloid beta was identified in spinal fluid as being a biological marker for AD long before memory problems become apparent. But testing spinal fluid is expensive and highly invasive, so AI was employed to identify proteins in the blood that can predict the concentration of amyloid beta in spinal fluid. The models they developed may help clinicians identify the risk of AD with up to 77 percent accuracy using a simple blood test.

This is the first study that uses AI to identify spinal fluid biomarkers through blood proteins, and the method can be easily extended to other key indicators of neurodegenerative disease. Such tests will allow researchers to more clearly identify and track onset of these conditions, enabling the development of more effective treatments.