A Cox-based Risk Prediction Model for Early Detection of Cardiovascular Disease

Jia, Xiaona
Mirza, Farhaan
GholamHosseini, Hamid
Baig, Mirza
Item type
Degree name
Master of Computer and Information Sciences
Journal Title
Journal ISSN
Volume Title
Auckland University of Technology

Cardiovascular disease (CVD) is the number one cause of mortality around the world. A fair proportion of health care resource is consumed for managing CVD, which imposes a heavy health burden on the community. To prevent the prevalence of CVD, an effective approach is to create prediction models to assess the CVD risk and then enable early lifestyle adjustments or clinical treatments. A great amount of research has been done but challenges and issues still exist. The aim of this research is to create an effective risk prediction model for early assessment and detection of CVD. The Framingham Original Cohort Study data set of 5079 subjects aged from 30 to 74 years old who had not previous symptoms of CVD at the baseline was enclosed. The Cox regression method was used for the data analysis. A complete process of creating risk models was conducted according to statistical regression strategies. Lastly, a risk prediction model for general CVDs was generated based on risk predictors, including age, sex, body mass index, hypertension, pulse rate, systolic blood pressure, cigarettes per day, and diabetes. We obtained a good predictive ability of discrimination and calibration with ROC of 0.71 indicating a good accuracy for the risk estimate of CVD. In our new Cox-based risk model, a novel predictor heart rate was incorporated to predict CVD risk, which expands the predictive ability of existing CVD risk models. Moreover, this risk prediction model was developed based on office risk factors, i.e. the measure of risk factors does not require clinical tests, which would be beneficial to both health care providers and patients to assess CVD event rates at any time and any place.

Cardiovascular disease (CVD) , Risk Prediction Model , Risk Factors , Cox Regression Model
Publisher's version
Rights statement