Evaluation of Machine Learning Methods to Predict Coronary Artery Disease Using Metabolomic Data

TL;DR

This study compares traditional regression methods and supervised machine learning algorithms like L1 regression and random forest for predicting coronary artery disease from metabolomic data, aiming to improve accuracy and exploit data complexity.

Contribution

It systematically evaluates machine learning methods against traditional regression for disease prediction using metabolomic data, highlighting their potential advantages.

Findings

Supervised machine learning methods outperform traditional regression in prediction accuracy.

Random forest achieves higher accuracy than L1 regression and traditional methods.

Machine learning approaches better capture complex metabolite interactions.

Abstract

Metabolomic data can potentially enable accurate, non-invasive and low-cost prediction of coronary artery disease. Regression-based analytical approaches however might fail to fully account for interactions between metabolites, rely on a priori selected input features and thus might suffer from poorer accuracy. Supervised machine learning methods can potentially be used in order to fully exploit the dimensionality and richness of the data. In this paper, we systematically implement and evaluate a set of supervised learning methods (L1 regression, random forest classifier) and compare them to traditional regression-based approaches for disease prediction using metabolomic data.