Exploring Dementia Risk in Cardiovascular Patients Through AI-Enabled Big Data Analysis

This study uses a large dataset from the UK Biobank to understand the link between cardiovascular diseases (CVDs) and increased risk of dementia. We analysed a dataset of over half a million patients. Data included 16043 variables on demographics, health status and genetic information. To the best of our knowledge, this is the largest study conducted with the UK population. To address the complexity and size of the data, we have developed a pipeline focused on data unpacking, prepro-cessing, descriptive and predictive analysis, and interpretation. Using post-training interpretation techniques, the study identified the top 40 variables to predict risk of dementia in people with CVD. For our predictive models, we employed support vector machines, logistic regression and random forest classifiers. The study revealed a strong correlation between CVD and dementia in the UK Biobank cohort, highlighting that individuals with CVD are at an elevated risk of developing dementia, indicated by a higher recall score of 0.89, in comparison to the overall population recall score of 0.82 for dementia.