It is now possible to predict the onset of Alzheimer’s disease up to seven years before symptoms appear, according to a new study by researchers at the University of California, San Francisco (UCSF). They developed an AI method that analyzes patient records to find specific patterns. These features can then be used to navigate larger Data baseData base Genetics to determine risk factors for disease. ” This is a great example of how we can leverage patient data with machine learning to predict which patients are more likely to develop Alzheimer’s disease and to understand why this happens. said Marina Sirota, lead author of the study and associate professor at UCSF.
The researchers compared the clinical data of people affected by Alzheimer’s disease with healthy people, including 5 million patients. Using their method, they found they could identify with predictive power 72% of people who would develop the disease seven years before symptoms appeared.
Predictive factors for Alzheimer’s disease
Several factors includingHypertensionHypertension AndHypercholesterolemiaHypercholesterolemia, were predictive of disease—both in men and women. Osteoporosis was a particularly strong predictor in women, who were more likely than men to develop Alzheimer’s disease. ” It is this combination of diseases that allows our model to predict the onset of Alzheimer’s disease.said Alice Tang, a doctoral student in UCSF’s Sirota laboratory. Our finding that osteoporosis is a predictor for women highlights the biological interaction between bone health and risk. frenzyfrenzy. »
Ultimately, the researchers hope that this approach can be used for other diseases that suffer from latency DiagnosticDiagnosticsuch as lupus and endometriosis.