What is the clinical evidence from this study?

Study design: Cross-Sectional. Population: Coronary artery disease undergoing successful percutaneous coronary intervention (PCI) (n=110). Intervention: Combined model (magnetocardiography + LDL, cTnI, NT-proBNP) vs. Magnetocardiography (MCG) model alone. Primary outcome: Prediction of postoperative Angina Frequency (SAQ-AF) episodes (AUC) (95% CI 0.722-0.903).

What question did this study set out to answer?

The aim is to develop and validate a machine learning model using magnetocardiography to predict post-PCI angina risk.

April 24, 2026SHILAP Revista de lepidopterologíaOpen Access

Adding clinical biomarkers to ML-based MCG improves post-PCI angina risk stratification over MCG alone.

View Full Paper

Ask AI

Bookmark

Why the study?

Recurrent angina after PCI impairs quality of life, but the role of magnetocardiography in predicting post-PCI symptomatic outcomes remains undefined.

Does a machine learning-based magnetocardiography model combined with clinical biomarkers predict post-PCI angina in patients with coronary artery disease?

Population

110 patients with coronary artery disease undergoing successful PCI

Comparison

Combined MCG and biomarker model vs MCG model alone

Follow-up

Within 3 months

Key result

Integrating a machine learning-based magnetocardiography model with clinical biomarkers predicted post-PCI angina frequency with an AUC of 0.813, improving risk stratification compared to MCG alone.

Authors

WWWenlong Wang LWL F Wang FDFaMing Ding

Discussion

Member takes

Overview

Combined MCG-clinical ML may refine post-PCI angina prediction; leaves open routine adoption pending larger prospective validation.

Key Points

The aim is to develop and validate a machine learning model using magnetocardiography to predict post-PCI angina risk.
Analyzed data from 110 patients who underwent successful PCI.
Conducted magnetocardiography before and after the procedure; measured clinical biomarkers.
Used multivariable logistic regression to create prediction models with performance evaluation metrics.
MCG scores dropped from 0.783 pre-PCI to 0.616 post-PCI.
The combined MCG and biomarker model performed better than MCG alone for SAQ outcomes.
The calibration of the prediction model showed satisfactory results.

Study Design

Type

Cross-Sectional (n=110)

Multicenter

Structured PICO

Does a machine learning-based magnetocardiography model combined with clinical biomarkers predict post-PCI angina in patients with coronary artery disease?

Population

110 patients with coronary artery disease undergoing successful percutaneous coronary intervention (PCI). Exclusions: age <18 years, NYHA functional class IV, bundle branch blocks, premature ventricular contractions, atrial fibrillation, other arrhythmias, or other systemic diseases.

Intervention

Machine learning-based magnetocardiography (MCG) model combined with clinical biomarkers (LDL-C, cTnI, NT-proBNP), with MCG performed pre-PCI and within 72 hours post-PCI.

Comparator

Machine learning-based magnetocardiography (MCG) model alone.

Outcome

Angina status assessed within 3 months using the Seattle Angina Questionnaire-Angina Stability (SAQ-AS) and -Angina Frequency (SAQ-AF) domains.patient reported

Main Result

Absolute Event Rate: 0.813% vs 0.801%

A machine learning-based magnetocardiography model, especially when combined with clinical biomarkers, effectively predicts the risk of recurrent angina after percutaneous coronary intervention.

Limitations

Single-center design coupled with a substantial number of excluded cases resulted in a relatively small final cohort
Advanced MCG system employed is currently confined to specialized centers
Machine learning model was trained on multi-institutional datasets, and inherent biases could affect performance
Follow-up period was limited to one year
Assessment of angina was based on the SAQ, which remains subjective and potentially influenced by non-cardiac factors

Cite This Study

Wang et al. (2026) conducted a cross-sectional in Coronary artery disease undergoing successful percutaneous coronary intervention (PCI) (n=110). Combined model (magnetocardiography + LDL, cTnI, NT-proBNP) vs. Magnetocardiography (MCG) model alone was evaluated on Prediction of postoperative Angina Frequency (SAQ-AF) episodes (AUC) (95% CI 0.722-0.903). Integrating a machine learning-based magnetocardiography model with clinical biomarkers predicted post-PCI angina frequency with an AUC of 0.813, improving risk stratification compared to MCG alone.

synapsesocial.com/papers/69eb07a4553a5433e34b3292 https://doi.org/10.3389/fcvm.2026.1750482

View Full Paper

Ask AI

Bookmark

April 24, 2026SHILAP Revista de lepidopterologíaOpen Access

Adding clinical biomarkers to ML-based MCG improves post-PCI angina risk stratification over MCG alone.

View Full Paper

Ask AI

Bookmark

Why the study?

Recurrent angina after PCI impairs quality of life, but the role of magnetocardiography in predicting post-PCI symptomatic outcomes remains undefined.

Does a machine learning-based magnetocardiography model combined with clinical biomarkers predict post-PCI angina in patients with coronary artery disease?

Population

110 patients with coronary artery disease undergoing successful PCI

Comparison

Combined MCG and biomarker model vs MCG model alone

Follow-up

Within 3 months

Key result

Integrating a machine learning-based magnetocardiography model with clinical biomarkers predicted post-PCI angina frequency with an AUC of 0.813, improving risk stratification compared to MCG alone.

Authors

WWWenlong Wang LWL F Wang FDFaMing Ding

Discussion

Member takes

Overview

Combined MCG-clinical ML may refine post-PCI angina prediction; leaves open routine adoption pending larger prospective validation.

Key Points

The aim is to develop and validate a machine learning model using magnetocardiography to predict post-PCI angina risk.
Analyzed data from 110 patients who underwent successful PCI.
Conducted magnetocardiography before and after the procedure; measured clinical biomarkers.
Used multivariable logistic regression to create prediction models with performance evaluation metrics.
MCG scores dropped from 0.783 pre-PCI to 0.616 post-PCI.
The combined MCG and biomarker model performed better than MCG alone for SAQ outcomes.
The calibration of the prediction model showed satisfactory results.

Study Design

Type

Cross-Sectional (n=110)

Multicenter

Structured PICO

Does a machine learning-based magnetocardiography model combined with clinical biomarkers predict post-PCI angina in patients with coronary artery disease?

Population

Intervention

Machine learning-based magnetocardiography (MCG) model combined with clinical biomarkers (LDL-C, cTnI, NT-proBNP), with MCG performed pre-PCI and within 72 hours post-PCI.

Comparator

Machine learning-based magnetocardiography (MCG) model alone.

Outcome

Angina status assessed within 3 months using the Seattle Angina Questionnaire-Angina Stability (SAQ-AS) and -Angina Frequency (SAQ-AF) domains.patient reported

Main Result

Absolute Event Rate: 0.813% vs 0.801%

A machine learning-based magnetocardiography model, especially when combined with clinical biomarkers, effectively predicts the risk of recurrent angina after percutaneous coronary intervention.

Limitations

Single-center design coupled with a substantial number of excluded cases resulted in a relatively small final cohort
Advanced MCG system employed is currently confined to specialized centers
Machine learning model was trained on multi-institutional datasets, and inherent biases could affect performance
Follow-up period was limited to one year
Assessment of angina was based on the SAQ, which remains subjective and potentially influenced by non-cardiac factors

Cite This Study

synapsesocial.com/papers/69eb07a4553a5433e34b3292 https://doi.org/10.3389/fcvm.2026.1750482

Also Consider

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

View Full Paper

Ask AI

Bookmark