Related Experiment Video
Updated: Jun 28, 2025
Quantification of Mouse Heart Left Ventricular Function, Myocardial Strain, and Hemodynamic Forces by Cardiovascular Magnetic Resonance Imaging
Published on: May 24, 2021
The role of beat-by-beat cardiac features in machine learning classification of ischemic heart disease (IHD) in
S Senthilnathan1, S Shenbaga Devi2, M Sasikala2
1SQUIDs Applications Section, SQUID and Detector Technology Division, Materials Science Group, Indira Gandhi Centre for Atomic Research, Kalpakkam-603 102, Tamil Nadu, India.
Beat-by-beat analysis of cardiac magnetic fields using magnetocardiography (MCG) significantly improves the detection of ischemic heart disease (IHD). This novel approach enhances classification accuracy compared to traditional methods, aiding in early diagnosis.
Area of Science:
- Cardiology
- Biophysics
- Machine Learning
Background:
- Ischemic heart disease (IHD) diagnosis can be challenging due to subtle cardiac electrical changes.
- Magnetocardiography (MCG) measures weak cardiac magnetic fields, offering potential for early detection.
- Previous machine learning (ML) studies often overlooked beat-by-beat variations, focusing on averaged cardiac signals.
Purpose of the Study:
- To investigate the utility of beat-by-beat features derived from MCG for classifying IHD subjects and healthy controls.
- To compare the classification performance of beat-by-beat features against conventional time-domain indices.
- To assess the broader applicability of beat-by-beat features in diagnosing myocardial infarction (MI).
Main Methods:
- Utilized 37-channel MCG data recorded under rest conditions from 23 IHD patients and 75 healthy controls.
- Extracted and analyzed eight beat-by-beat features, including field map angle (FMA) and ST-T region variations.
- Employed ML classifiers to distinguish between IHD and control groups, and validated findings on public ECG databases for MI classification.
Main Results:
- Beat-by-beat features achieved a superior classification accuracy of 92.7% for IHD detection, compared to 81% with conventional features.
- Identified three key beat-by-beat features (FMA, alpha angle variations, T wave magnitude variations) as crucial for improved classification.
- Demonstrated enhanced accuracy in classifying myocardial infarction (MI) versus control subjects using public ECG data (92% vs. 88% and 94% vs. 77%).
Conclusions:
- Beat-by-beat feature analysis in MCG is highly effective for improving the accuracy of ischemic heart disease detection.
- This approach offers a significant advancement over traditional methods that rely on averaged cardiac signals.
- The findings underscore the clinical importance of beat-by-beat variations for objective and accurate ischemia diagnosis.
