Related Experiment Video
Updated: Nov 10, 2025
Asthma Detection Research Based on Voice Signal Processing and Machine Learning
Published on: July 22, 2025
Automatic Premature Ventricular Contraction Detection Using Deep Metric Learning and KNN
Junsheng Yu1, Xiangqing Wang1, Xiaodong Chen2
1School of Electronic Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China.
Insights
This study introduces a novel, automated method for detecting premature ventricular contractions (PVCs) from long-term ECGs using deep metric learning and KNN classification, achieving high accuracy without complex preprocessing.
Area of Science:
- Cardiology
- Biomedical Engineering
- Artificial Intelligence in Medicine
Background:
- Premature ventricular contractions (PVCs) are common irregular heartbeats indicating potential heart disease.
- Long-term electrocardiograms (ECG) from wearable devices are crucial for PVC diagnosis but analysis is time-consuming.
- Current methods often rely on manual feature engineering, introducing potential bias.
Purpose of the Study:
- To develop a simplified, automated approach for detecting PVCs from long-term ECG data.
- To leverage deep metric learning for intelligent, bias-free feature extraction.
- To improve the efficiency and accuracy of PVC detection.
Main Methods:
- Utilized deep metric learning for supervised feature extraction, focusing on intra-product variance and inter-product differences.
- Employed a k-nearest neighbors (KNN) classifier to detect PVCs based on extracted heartbeat features.
- Evaluated the method on the MIT-BIH Arrhythmia Database without complex preprocessing.
Main Results:
- Achieved high diagnostic performance: 99.7% accuracy, 97.45% sensitivity, and 99.87% specificity.
- Demonstrated the reliability of deep metric learning and KNN for PVC recognition.
- Confirmed the method's effectiveness without requiring complicated preprocessing steps.
Conclusions:
- The proposed deep metric learning and KNN approach offers a reliable and automated solution for PVC detection from long-term ECGs.
- This method overcomes limitations of manual feature engineering, reducing bias and improving efficiency.
- The technique shows significant potential for clinical application in diagnosing irregular heartbeats.
Abstract:
Premature ventricular contractions (PVCs), common in the general and patient population, are irregular heartbeats that indicate potential heart diseases. Clinically, long-term electrocardiograms (ECG) collected from the wearable device is a non-invasive and inexpensive tool widely used to diagnose PVCs by physicians. However, analyzing these long-term ECG is time-consuming and labor-intensive for cardiologists. Therefore, this paper proposed a simplistic but powerful approach to detect PVC from long-term ECG. The suggested method utilized deep metric learning to extract features, with compact intra-product variance and separated inter-product differences, from the heartbeat. Subsequently, the k-nearest neighbors (KNN) classifier calculated the distance between samples based on these features to detect PVC. Unlike previous systems used to detect PVC, the proposed process can intelligently and automatically extract features by supervised deep metric learning, which can avoid the bias caused by manual feature engineering. As a generally available set of standard test material, the MIT-BIH (Massachusetts Institute of Technology-Beth Israel Hospital) Arrhythmia Database is used to evaluate the proposed method, and the experiment takes 99.7% accuracy, 97.45% sensitivity, and 99.87% specificity. The simulation events show that it is reliable to use deep metric learning and KNN for PVC recognition. More importantly, the overall way does not rely on complicated and cumbersome preprocessing.
