Related Experiment Video
Updated: Jan 28, 2026

Measurement of the Directional Information Flow in fNIRS-Hyperscanning Data using the Partial Wavelet Transform Coherence Method
Published on: September 3, 2021
A Wavelet Transform-Based Neural Network Denoising Algorithm for Mobile Phonocardiography.
Dawid Gradolewski1, Giovanni Magenes2, Sven Johansson3
1Blekinge Institute of Technology, Institute of Applied Signal Processing, 371 79 Karlskrona, Sweden. dawid.gradolewski@bth.se.
This study introduces an adaptive denoising algorithm combining Wavelet Transform (WT) and Time Delay Neural Networks (TDNN) to improve the accuracy of phonocardiography (PCG) for early cardiac illness detection.
Area of Science:
- Biomedical Engineering
- Signal Processing
- Cardiology
Background:
- Cardiovascular diseases are a leading cause of death globally.
- Phonocardiography (PCG) is a valuable tool for diagnosing heart conditions.
- PCG signals often require denoising due to environmental and physiological noise.
Purpose of the Study:
- To develop an adaptive denoising algorithm for enhancing phonocardiography (PCG) signals.
- To improve the early detection of cardiovascular pathologies.
- To overcome limitations of traditional Wavelet Transform (WT) denoising methods.
Main Methods:
- A novel adaptive denoising algorithm combining WT and Time Delay Neural Networks (TDNN).
- WT decomposition using the coif five-wavelet basis at the tenth level.
- TDNN with optimized parameters (25 neurons, 15 neurons, 12-sample delay block) for adaptive thresholding and Inverse Wavelet Transform (IWT) estimation.
Main Results:
- The proposed algorithm effectively denoises pathological heart sounds and signals from noisy environments.
- The combined WT-TDNN approach demonstrates superior performance compared to existing WT-based denoising methods.
- Validation through an online questionnaire confirmed the system's effectiveness.
Conclusions:
- The adaptive WT-TDNN algorithm offers a robust solution for PCG signal enhancement.
- This method has the potential to significantly aid in the early diagnosis of cardiovascular diseases.
- Further development could lead to advanced auto-diagnostic systems for heart monitoring.
Related Concept Videos
Protein Networks
These interactions can be represented through maps depicting protein-protein interaction networks, represented as nodes and edges. Nodes are circles that are representative of a protein,...
Bacterial Transformation
Griffith made an unexpected discovery when he killed the pathogenic strain and mixed its remains with the live, non-pathogenic strain. Not only did the mixture kill host mice, but it also contained living pathogenic bacteria that...
Trial and Error and Algorithm
Network Covalent Solids
To break or to melt a covalent network solid, covalent bonds must be broken. Because covalent bonds are relatively strong, covalent network solids are typically...
Transformers
The iron core has a substantial relative permeability. Therefore, the magnetic field lines generated due to the current in one winding are almost entirely confined within the core, such that the same magnetic flux permeates each turn of both...
Transformation

