An XGBoost risk prediction model of cardiovascular and cerebrovascular diseases with plateau healthcare dataset

Abstract

This paper aims to build an XGBoost risk prediction model of cardiovascular and cerebrovascular diseases (CVDs) with plateau healthcare dataset. The incidence of cardiovascular and cerebrovascular diseases is very high in plateau areas. And it is difficult to detect partly due to the high cost of professional test. It will have high practical value to build a model to predict the risk of cardiovascular and cerebrovascular diseases by using the common healthcare data (e.g. fundus data). The paper proposes an XGBoost prediction model of CVDs risk with the fundus disease and other healthcare data set. The influence of various fundus disease factors on cardiovascular and cerebrovascular diseases is analyzed in the study. The result suggests that the proposed XGBoost prediction model performs better in terms of accuracy and recall rate compared with other models.