Key result
The proposed wavelet-based Convolutional Neural Network using PPG and ABP signals classified Premature Ventricular Contractions into Normal, P1, and P2 types with accuracies of 99.9%, 99.8%, and 99.5%.
Why the study?
Accurate and precise monitoring of cardiac arrhythmias helps to avoid serious heart issues.
A deep convolutional neural network utilizing wavelet transforms on PPG and ABP signals can accurately detect and classify premature ventricular contractions, offering a potential alternative to ECG-based monitoring.
Accurate and precise monitoring of cardiac arrhythmia’s helps to avoid serious heart issues. The research concentrates on using Photoplethysmography (PPG) and Arterial Blood Pressure (ABP) with deep Convolutional Neural Networks (CNN) for the classification and detection of fetal cardiac arrhythmia or Premature Ventricular Contractions (PMVCs). The process starts with Icentia 11k, a public dataset of ECG signal which consist of different cardiac abnormalities. The process proceeds with the MIMIC dataset and test dataset available from the hospital which is the transferred weights obtained from Icentia 11k dataset. The fine tuning improves the accuracy of classification. The proposed method can be able to detect and classify PMVCs in to three types: Normal, P1 and P2 with an accuracy of 99.9%, 99.8% and 99.5%.
No takes yet. Share an insight, caveat, or question.
Raj et al. (2022) studied Premature Ventricular Contraction (PMVC) (n=11,050). Wavelet-based Convolutional Neural Network (CNN) vs. Standard machine learning classifiers was evaluated on Classification accuracy for Normal, P1, and P2 pulses. The proposed wavelet-based Convolutional Neural Network using PPG and ABP signals classified Premature Ventricular Contractions into Normal, P1, and P2 types with accuracies of 99.9%, 99.8%, and 99.5%.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: