The use of cluster analysis and logistic regression for assessing the risk of Alzheimer's disease in patients with mild cognitive impairment, amnestic type

Abstract

The evaluation of the risk of Alzheimer disease (AD) in patients with cognitive impairment, amnestic type (aMCI) on the basis of cluster analysis and logistic regression with the use of such markers of inflammation as enzymatic activity of leukocyte elastase (LE) and the functional activity of α1-proteinase inhibitor (α1-PI). The study object of statistical analysis was the database, including the results of LE activity and functional α1-PI activity in blood plasma of 78 outpatients with aMCI (25 men and 53 women, aged 44 to 89 years (69.1±9.95). Clustering by k-means and classification by logistic regression indicate a high probability of AD in patients with aMCI depending on the activity of LE and α1-PI in blood plasma. The total coincidence of objects included in the clusters and in the AD risk group was 94%. The high coincidence of two different methods of grouping confirms the previously stated notion of the possibility of identifying patients with the high risk of AD among patients with aMCI by the activity of LE and α1-PI in the blood.