Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 16, 2025Physiological MeasurementOpen Access

Machine learning with OPM-MCG features yields ~0.86 AUC for CAD diagnosis, outperforming clinical features alone.

View Full Paper
Ask AI
Bookmark
Share

Why the study?

The potential of optical pumped magnetometer magnetocardiography (OPM-MCG) for diagnosing coronary artery disease had been initially shown, but large-scale prospective research was lacking.

Does machine learning using optical pumped magnetometer magnetocardiography (OPM-MCG) accurately diagnose coronary artery disease compared to invasive coronary angiography?

Population

1513 participants evaluated for CAD with ICA reference

Comparison

OPM-MCG Heart features vs Clinical features vs Heart plus Clinical combined feature sets across 11 ML models

Design

Prospective cohort study

Key result

Machine learning models using optical pumped magnetometer magnetocardiography features achieved high diagnostic accuracy for coronary artery disease (AUC 0.84–0.88), outperforming clinical features alone.

Authors

CTChenchen TuCapital Medical UniversitySYShuwen YangChinese PLA General HospitalZWZhixiang WangChina Pharmaceutical University

Discussion

Loading...

Member takes

Overview

May support noninvasive CAD evaluation in cohorts; leaves open need for prospective validation before clinical adoption.

Study Design

Type

Cohort (n=1,513)

Structured PICO

Does machine learning using optical pumped magnetometer magnetocardiography (OPM-MCG) accurately diagnose coronary artery disease compared to invasive coronary angiography?

P
Population
1,513 participants (mean age 58.2 years, 24.5% female) evaluated for coronary artery disease using invasive coronary angiography as a reference.
E
Exposure
Machine learning models utilizing optical pumped magnetometer magnetocardiography (OPM-MCG) features ('Heart' features) and/or 'Clinical' features.
C
Comparator
Invasive coronary angiography (ICA) as the reference standard.
O
Outcome
Diagnostic accuracy (Area Under the Curve, AUC) for detecting coronary artery disease.surrogate

Main Result

Effect estimate: AUC 0.84-0.88

Machine learning models based on optical pumped magnetometer magnetocardiography (OPM-MCG) demonstrate high diagnostic accuracy for coronary artery disease, outperforming clinical features alone.

Cite This Study

Tu et al. (2025) conducted a cohort in Coronary artery disease (n=1,513). Optical pumped magnetometer magnetocardiography (OPM-MCG) with machine learning vs. Clinical features alone was evaluated on Diagnostic accuracy for coronary artery disease (AUC 0.84-0.88). Machine learning models using optical pumped magnetometer magnetocardiography features achieved high diagnostic accuracy for coronary artery disease (AUC 0.84–0.88), outperforming clinical features alone.

synapsesocial.com/papers/6a2170f1df884daff757eec8https://doi.org/10.1088/1361-6579/adf0be
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Noninvasive diagnosis of ischemia non-obstructive coronary using machine learning-assisted magnetocardiography2025
  2. 2Magnetocardiography-based coronary artery disease severity assessment and localization using spatiotemporal features2023 · 21 citations
  3. 3Machine Learning Application in Different Imaging Modalities for Detection of Obstructive Coronary Artery Disease and Outcome Prediction: A Systematic Review and Meta-Analysis2025 · 1 citations
  4. 4Magnetocardiography for Distinguishing Obstructive Coronary Heart Disease and Coronary Microvascular Dysfunction: A Feasibility Study2026 · 2 citations
  5. 5Machine learning models of clinically relevant biomarkers for the prediction of stable obstructive coronary artery disease2022 · 14 citations