Pre-existing and machine learning-based models for cardiovascular risk prediction

Sang-Yeong Cho,Tackeun Kim,Kyong Joon Lee,In-Ho Chae,Chang-Hwan Yoon,Sun-Hwa Kim,Seungjin Kang,Sang Jun Park,Si-Hyuck Kang,Tae-Jin Youn,Dongjun Choi

doi:10.1038/s41598-021-88257-w

Abstract

Predicting the risk of cardiovascular disease is the key to primary prevention. Machine learning has attracted attention in analyzing increasingly large, complex healthcare data. We assessed discrimination and calibration of pre-existing cardiovascular risk prediction models and developed machine learning-based prediction algorithms. This study included 222,998 Korean adults aged 40–79 years, naïve to lipid-lowering therapy, had no history of cardiovascular disease. Pre-existing models showed moderate to good discrimination in predicting future cardiovascular events (C-statistics 0.70–0.80). Pooled cohort equation (PCE) specifically showed C-statistics of 0.738. Among other machine learning models such as logistic regression, treebag, random forest, and adaboost, the neural network model showed the greatest C-statistic (0.751), which was significantly higher than that for PCE. It also showed improved agreement between the predicted risk and observed outcomes (Hosmer–Lemeshow χ2 = 86.1, P < 0.001) than PCE for whites did (Hosmer–Lemeshow χ2 = 171.1, P < 0.001). Similar improvements were observed for Framingham risk score, systematic coronary risk evaluation, and QRISK3. This study demonstrated that machine learning-based algorithms could improve performance in cardiovascular risk prediction over contemporary cardiovascular risk models in statin-naïve healthy Korean adults without cardiovascular disease. The model can be easily adopted for risk assessment and clinical decision making.

Highlights

Predicting the risk of cardiovascular disease is the key to primary prevention
5-year event rates varied from 0.30% in the systematic coronary risk evaluation (SCORE) cohort-where only cardiac death was counted to 3.51% in the Pooled cohort equation (PCE) cohort where hard atherosclerotic Cardiovascular disease (CVD) was counted
We found that pre-existing risk models showed acceptable performance in predicting cardiovascular risk in real-world Korean adults who were free from CVD and naïve to statin therapy

Summary

Introduction

Predicting the risk of cardiovascular disease is the key to primary prevention. Machine learning has attracted attention in analyzing increasingly large, complex healthcare data. This study demonstrated that machine learning-based algorithms could improve performance in cardiovascular risk prediction over contemporary cardiovascular risk models in statin-naïve healthy Korean adults without cardiovascular disease. Abbreviations ACC/AHA American College of Cardiology /American Heart Association AUC Area under curve CIs Confidence intervals CVD Cardiovascular disease FRS Framingham risk score ICD‐10 International Classification of Diseases, 10th Revision MESA Multi-Ethnic Study of Atherosclerosis ML Machine learning NHIS-HEALS National Health Insurance Service-Health Screening PCE Pooled cohort equation SCORE Systematic coronary risk evaluation. ML methods have been increasingly applied in imaging interpretation and shown promising results[15, 16] They can be used to develop prediction models from existing data to yield highly accurate r esults[17]

Methods

Results

Conclusion