CardioDB: A Relational Database and Knowledge Graph of Cardiovascular Disease

Abstract

Background: Cardiovascular disease (CVD) poses a significant and persistent challenge to global public health. Worldwide, heart attacks and strokes account for 80% of cardiovascular disease fatalities, many of which can be averted with proactive lifestyle modifications and timely medical interventions. Given the significant mortality rates, the clinical health sector is producing substantial volumes of data. These healthcare sectors face significant challenges in managing multidimensional data, encompassing everything from handwritten notes to high-resolution images to medical images. Consequently, an effective database is required to address the challenges of data storage and accessibility. Method: This database consists clinical records of patients with CVD, with the highest number of feature parameters recorded. In addition, a CVD calculator was developed to analyze user risk based on 13 major clinical parameters, and the risk was measured via the weighted assisted scoring (WAS) function. Results: The database features a user-friendly web interface that allows users to easily search, navigate, and discover knowledge. This database was designed to increase accessibility, support researchers in addressing high-end feature profiles, improve treatment decision-making, enable time-series pattern analysis, and provide health professionals and researchers with specific reports with current patient information to refine future diagnostic accuracy.