Breakthrough in Early Detection of Coronary Artery Disease with High-Sensitivity MEMS-Based Technology

Strategic Insights -

Research from the North University of China has presented a significant advancement in early detection of coronary artery disease (CAD) using a high-sensitivity MEMS-based PCG-ECG Synchronous Auscultation System. This system has been developed to construct a high-fidelity clinical dataset of synchronous PCG-ECG data for CAD, leveraging deep learning technology to extract multi-resolution features and achieve high-precision detection. The research has reported a high detection accuracy of 96.29% on the clinical dataset and 97.69% on the public dataset.

Key Takeaways:

  • Researchers from the North University of China have developed a high-sensitivity MEMS-based PCG-ECG Synchronous Auscultation System for early detection of coronary artery disease (CAD).
  • The system uses a Multi-scale Cross-modal Attention Fusion Network (MCAF-Net) to extract multi-resolution features from synchronous PCG-ECG spectrograms and achieve high-precision detection.
  • The research has reported a detection accuracy of 96.29% on the clinical dataset and 97.69% on the public dataset.
  • The high-precision detection of CAD is crucial for reducing its mortality rate, highlighting the importance of this research in healthcare.
  • The research has the potential to revolutionize early detection and treatment of CAD, saving countless lives worldwide.
  • The study's authors, Chen Fangyuan et al., are affiliated with the State key Laboratory of Extreme Environment Optoelectronic Dynamic Testing Technology and Instrument, North University of China.
  • The research has been published in the MATEC Web of Conferences, 2025, 413():01007.

Statistics:

  • 96.29% detection accuracy on the clinical dataset
  • 97.69% detection accuracy on the public dataset
  • Coronary artery disease (CAD) is one of the leading causes of death worldwide
  • High-resolution features extracted from synchronous PCG-ECG spectrograms using an improved residual network
  • Feature interaction and fusion through Mutual Cross Attention (MCA) achieving high-precision detection

Sources:

  • MATEC Web of Conferences, 2025, 413():01007
  • https://doi-org.sdpl.idm.oclc.org/10.1051/matecconf/202541301007
  • NewsRx LLC, 2025