Prediction of Alzheimer’s Disease Risk Based on Plasma Lipid Profiles and Covid-19 History Via Logistic Regression.

Abstract

Alzheimer’s disease (AD), a form of dementia is known as a neurodegenerative disorder with gradual memory impairment. Recent research studies have suggested a putative risk of increased susceptibility to AD post COVID-19 infection. This study attempted to clarify this relationship by examining lipid profile data in conjunction with the information available in patient database of COVID-19. Data were analyzed by preprocessing, exploratory data analysis through pivot tables and a logistic regression model for the prediction of risk of AD. The model adjusted for APOE4 genotype, age and COVID-19 status. Model performance metrics such as accuracy, precision, recall and F1 score were evaluated. Although reasonably accurate, there was a high rate of false negatives in the model and more work is required to improve these. These results emphasized the intricate association of lipid profiles, COVID-19 infection and AD risk. Identification of biomarkers and effective risk assessment strategy for AD in the setting of COVID-19 needs more investigations. The timely diagnosis and intervention may lead to focused therapy and preventive interventions that reduce the chronic neurological impact of COVID-19.