PCG Signal Analysis with Adaptive Filtering

Project Overview

1

What is PCG?

PCG stands for Phonocardiogram. It is a recording of the sounds made by the heart during its functioning.

2

Importance of PCG Signal Analysis

PCG signal analysis is crucial for several reasons:

3

Brief Explanation of Adaptive Filtering

Adaptive filtering is a signal processing technique used to remove noise and unwanted components from a signal.

In this project, adaptive filtering is applied to PCG signals to enhance their quality, potentially leading to more accurate heart sound analysis and diagnosis.

Key Features

• Multi-stage Feed-forward Adaptive Filter

  • Utilizes multiple stages of filtering for enhanced noise reduction
  • Each stage builds upon the improvements made by the previous stage

• Automatic Adjustment of Filter Stages

  • The system can dynamically adapt the number of filter stages based on signal quality
  • Optimizes computational efficiency while maintaining high performance

• Adaptive Algorithm Switching

  • Alternates between Least Mean Squares (LMS) and Sign-LMS algorithms
  • Balances between convergence speed and stability for optimal performance

• Configurable Filter Parameters

  • Adjustable filter length for fine-tuning the trade-off between noise reduction and computational complexity
  • Customizable step size (learning rate) to control adaptation speed and stability

• Synthetic PCG Signal Generation

  • Includes capability to generate synthetic PCG signals for testing and validation
  • Allows for controlled experiments with known signal characteristics

• Customizable Noise Addition

  • Can add controlled amounts of noise to clean signals
  • Enables testing of the filter's performance under various noise conditions

• Real-time Signal-to-Noise Ratio (SNR) Calculation

  • Continuously measures SNR improvement
  • Provides quantitative assessment of filter performance

• Visualizations of Signal Processing Stages

  • Generates plots of original, noisy, and filtered signals
  • Enables visual inspection and comparison of filter effectiveness

• Flexible Input Handling

  • Can process both synthetic and real PCG recordings
  • Adaptable to various PCG signal formats and sampling rates

• Performance Metrics Reporting

  • Calculates and reports key performance indicators like SNR improvement
  • Facilitates objective evaluation and comparison of different filter configurations

• Modular Design

  • Separates signal generation, noise addition, filtering, and analysis components
  • Allows for easy extension and modification of individual components

• Scalable Processing

  • Capable of handling both short PCG segments and longer recordings
  • Suitable for both quick tests and extended monitoring scenarios

• Error Analysis

  • Computes and tracks error signals throughout the filtering process
  • Useful for debugging and fine-tuning filter performance

• Adaptive to Various Noise Types

  • Effective against different noise sources common in PCG recordings (e.g., respiratory sounds, ambient noise)
  • Robust performance across diverse recording conditions

Reports

Project Reports and Documentation

Reports and documentation related to the project are available for download. Click the button below to view and download the reports. (Includes presentations, reports, and more)

View Reports

Resources

View our resources used in the project

References and resources used in the project are available for further reading. Click the button below to view the resources. (Includes research papers, articles, and more)

View Resources