What is the clinical evidence from this study?

Study design: Review. Population: Acute myocardial infarction. Intervention: Machine learning methods vs. Regression models. Primary outcome: Mortality.

What question did this study set out to answer?

This review aims to demonstrate the potential of machine learning methods for improving risk prediction models in cardiovascular settings.

April 21, 2026Open Access

Moving beyond regression techniques in cardiovascular risk prediction: applying machine learning to address analytic challenges

Key Result

Machine learning methods provide alternative approaches to traditional regression models for addressing complex analytic challenges in cardiovascular risk prediction.

Key Points

This review aims to demonstrate the potential of machine learning methods for improving risk prediction models in cardiovascular settings.
Review traditional regression models and their limitations in risk prediction.
Utilize data from electronic health records and laboratory markers for analysis.
Introduce various machine-learning approaches to address specific analytic challenges.
Machine-learning methods provide better flexibility in handling predictors across varied populations.
Challenges of traditional regression, including limited predictors and uniformity, are effectively tackled.
Discussion highlights the importance of tuning parameters and handling missing data in machine learning.

Structured PICO

How can machine learning methods address analytic challenges in cardiovascular risk prediction compared to traditional regression techniques?

Population

Patients with acute myocardial infarction (used as an illustrative example derived from electronic health records)

Intervention

Machine learning methods for risk prediction model development

Comparator

Traditional regression models

Outcome

Mortality after diagnosis of acute myocardial infarction (illustrative)

This review provides an introduction to applying machine learning techniques to overcome the limitations of traditional regression models in cardiovascular risk prediction.

Limitations

Machine learning methods are typically presented as black box approaches
Application issues include tuning parameters, loss functions, variable importance, and missing data

Abstract

Risk prediction plays an important role in clinical cardiology research. Traditionally, most risk models have been based on regression models. While useful and robust, these statistical methods are limited to using a small number of predictors which operate in the same way on everyone, and uniformly throughout their range. The purpose of this review is to illustrate the use of machine-learning methods for development of risk prediction models. Typically presented as black box approaches, most machine-learning methods are aimed at solving particular challenges that arise in data analysis that are not well addressed by typical regression approaches. To illustrate these challenges, as well as how different methods can address them, we consider trying to predicting mortality after diagnosis of acute myocardial infarction. We use data derived from our institution's electronic health record and abstract data on 13 regularly measured laboratory markers. We walk through different challenges that arise in modelling these data and then introduce different machine-learning approaches. Finally, we discuss general issues in the application of machine-learning methods including tuning parameters, loss functions, variable importance, and missing data. Overall, this review serves as an introduction for those working on risk modelling to approach the diffuse field of machine learning.