MACHINE-LEARNING-BASED FORECASTING OF THE PROGRESSION OF ALZHEIMER'S DISEASE

摘要

According to one or more embodiments herein, a machine-learning algorithm is provided that uses current and past clinical data in order to accurately and precisely predict the future onset of mild cognitive impairment (MCI) and dementia for individual patients, thus enabling early identification of those having high risk for Alzheimer's disease. In particular, a newly defined “All-Pairs” technique combines data from each doctor's visit for each patient with data from each of the patient's other doctor's visits. By correlating clinical data obtained from patients at one time point with the progression of Alzheimer's disease in the future, the techniques herein are able to increase the likelihood of identifying Alzheimer's disease patients at early stages.