Related Experiment Video
Updated: Jun 3, 2025
Coronary Progenitor Cells and Soluble Biomarkers in Cardiovascular Prognosis after Coronary Angioplasty
Published on: January 28, 2020
Evaluation of a machine learning-based metabolic marker for coronary artery disease in the UK Biobank
Kyle Gibson1, Iain S Forrest2, Ben O Petrazzini3
1The Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai, New York, NY, USA; Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, New York, NY, USA.
Insights
A new metabolic risk score for coronary artery disease (CAD) effectively predicts atherosclerosis, complications, and mortality. This machine learning approach using metabolomic data offers a powerful tool for assessing cardiovascular risk.
Area of Science:
- Cardiovascular Disease Research
- Machine Learning in Medicine
- Metabolomics
Background:
- Existing in silico scores for coronary artery disease (CAD) use clinical data but haven't fully leveraged metabolomic profiling.
- Metabolite biomarker profiling offers enhanced portability and objectivity for machine learning in disease prediction.
Purpose of the Study:
- To evaluate a quantitative CAD risk score derived from a machine learning model trained on metabolomic data.
- To assess the association of this metabolic CAD risk score (M-CAD) with indicators of atherosclerosis, CAD sequelae, and mortality.
Main Methods:
- Developed a CAD-predictive machine learning model using metabolic data from 93,642 UK Biobank participants.
- Assessed the model's probabilities as a quantitative metabolic risk score for CAD (M-CAD).
- Examined the relationship of M-CAD with arterial stiffness index, ejection fraction, CAD sequelae, and mortality.
Main Results:
- The model achieved an area under the ROC curve of 0.712 for CAD prediction.
- Increased M-CAD correlated with higher arterial stiffness and lower ejection fraction.
- Higher M-CAD quartiles/deciles showed stepwise increases in myocardial infarction, all-cause mortality, CVD-associated mortality, and CAD-associated mortality.
Conclusions:
- A metabolic-based machine learning model can generate a quantitative CAD risk score.
- This M-CAD score is significantly associated with atherosclerotic burden, CAD complications, and mortality risk.
Background And Aims:
An in silico quantitative score of coronary artery disease (ISCAD), built using machine learning and clinical data from electronic health records, has been shown to result in gradations of risk of subclinical atherosclerosis, coronary artery disease (CAD) sequelae, and mortality. Large-scale metabolite biomarker profiling provides increased portability and objectivity in machine learning for disease prediction and gradation. However, these models have not been fully leveraged. We evaluated a quantitative score of CAD derived from probabilities of a machine learning model trained on metabolomic data.
Methods:
We developed a CAD-predictive learning model using metabolic data from 93,642 individuals from the UK Biobank (median [IQR] age, 57 [14] years; 39,796 [42 %] male; 5640 [6 %] with diagnosed CAD), and assessed its probabilities as a quantitative metabolic risk score for CAD (M-CAD; range 0 [lowest probability] to 1 [highest probability]) in participants of the UK Biobank. The relationship of M-CAD with arterial stiffness index, ejection fraction, CAD sequelae, and mortality was assessed.
Results:
The model predicted CAD with an area under the receiver-operating-characteristic curve of 0.712. Arterial Stiffness Index increased by 0.19 and ejection fraction decreased by 0.2 % per 0.1 increase in M-CAD. Both incident and recurrent myocardial infarction increased stepwise over M-CAD quartiles (odds ratio (OR) 15.3 [4.2 %] and 12.5 [0.2 %]) in top quartiles as compared to the first quartile of incident and recurrent MI respectively). Likewise, the hazard ratio and prevalence of all-cause mortality, CVD-associated mortality, and CAD-associated mortality increased stepwise over M-CAD deciles (2.98 [14 %], 9.34 [4.3 %], 26.7 [2.7 %] in the top deciles as compared to the first decile of all-cause, CVD, and CAD mortality respectively).
Conclusions:
Metabolic-based machine learning can be used to build a quantitative risk score for CAD that is associated with atherosclerotic burden, CAD sequelae and mortality.
More Related Videos
Related Concept Videos
01:19Blood Studies for Cardiovascular System II: CRP, Hcy, and Cardiac Natriuretic Peptide Markers
These markers indicate stress or strain on the heart muscle:
Natriuretic Peptides (BNP)
Cardiac myocytes produce these hormones in response to ventricular stretching...
01:20Blood Studies for Cardiovascular System I: Cardiac Biomarkers
The essential diagnostic tools for detecting myocardial necrosis and monitoring individuals suspected of having acute coronary syndrome (ACS) include:
Troponins
Troponins, particularly cardiac troponins I and T, are the most precise and sensitive markers of myocardial injury. They are detectable within 4-6 hours of myocardial injury and remain...