The digital symbol digit test : screening for Alzheimer's and Parkinson's

Abstract

Neurodegenerative diseases affect the cognition of millions of people worldwide, degrading their quality of life and placing a burden on their families. Early identication can be extremely beneficial in treating or slowing down the onset of these diseases. One technique used to identify early warning signs is the use of cognitive tests. Unfortunately, grading these tests is subjective. In this study, we quantitatively evaluated the digital Symbol Digit Test (dSDT), in which patients translate symbols into digits based on a given mapping. In collaboration with Dr. Penney of Lahey Clinic, we administered the dSDT to over 170 patients using a digitizing pen that measures its position on the page and the pressure applied. We developed support vector machine and logistic regression classifiers that indicate Alzheimer's Disease and Parkinson's Disease with an area under the curve of 0.957 and 0.963, respectively.