REVOLUTIONIZING RISK ASSESSMENT: AN INTEGRATED BIOSTATISTICAL APPROACH TO CARDIO-CEREBROVASCULAR DISEASE RISK FACTORS IN WORKING-AGED ADULTS WITH SILENT CEREBRAL SMALL VESSEL DISEASE

Abstract

Age-related cerebral small vessel disease (CSVD) such as white matter hyperintensities (WMHs) are always recognised as an incidental finding following neuroimaging and associated with increased predisposition to cardio-cerebrovascular disease risk factors. This study aims to determine the most factor/s that may have the highest association with the prevalence of CSVD among working-aged adults. Sixty subjects (mean age: 39.83 ±11.50 years) with low to moderate cardio-cerebrovascular disease risk prediction scores based on QRISK3 were recruited and underwent 3T magnetic resonance imaging (MRI) brain scan and their peripheral venous blood was collected for enumeration of circulating microparticles (MPs). An integrated biostatistical approach using correlation analysis (CA), decision tree analysis (DTA), and feedforward artificial neural networks (ANN) i.e., multilayer perceptron (MLP) was used to determine the associated factor/s. The prevalence of CSVD was found in 20 subjects (mean age: 46.00 ± 12.00 years). CA plot reveals that age, hypertension, smoking status, and family history had a strong correlation to inform the CSVD status among subjects. DTA revealed that ageing and platelet-derived MPs (CD62P) are the most important predictors for CSVD. MLP model further confirms that ageing and CD62P are the most important cardio-cerebrovascular disease risk predictors for the prevalence of CSVD among the study population with 88.3% accuracy. This study also provides new insight into the level of MPs as a novel potential biomarker for CSVD and a benchmark for decision making especially among clinicians to improve the quality of assessment, prevention, and therapeutics strategies given to the working-aged adults with CSVD.