Assessing the Effectiveness of Various Filtering Techniques on Seismocardiography Signals in Individuals with Valvular Heart Disease
Executive Summary
This study evaluates the effectiveness of various digital filtering techniques, including ICA, PCA, Butterworth, Chebyshev, Wavelet Transform, EMD, VMD, and CWT, for denoising seismocardiography (SCG) signals in patients with valvular heart disease. Performance metrics such as SNR, PSNR, PRD, SSIM, and MSE were used to assess the filters, with ICA emerging as the most effective method for preserving signal integrity while reducing noise. The findings provide a systematic framework for selecting optimal filtering techniques tailored to specific diagnostic applications.
Answer Machine Insights
Q: Which filtering method performed best in preserving SCG signal integrity?
Independent Component Analysis (ICA) performed best, achieving an SSIM score of 1.00 and the highest SNR and PSNR values.
ICA was outstanding and was the only method that rendered a perfect SSIM score of 1.00 indicating that it is able to maintain the structural similarity of the SCG signal with respect to its original even after denoising.
Q: What metrics were used to evaluate the filtering techniques?
The metrics used were Signal-to-Noise Ratio (SNR), Peak Signal-to-Noise Ratio (PSNR), Peak Relative Difference (PRD), Structural Similarity Index (SSIM), and Mean Square Error (MSE).
Performance evaluation was done using SNR, PSNR, PRD, SSIM, and MSE.
Key Results
ICA achieved the highest SSIM score of 1.00, indicating perfect structural similarity preservation.
ICA demonstrated the highest SNR (23.21) and PSNR (47.57), confirming its superior noise reduction capabilities.
Research Tags
Clinical Snapshot
Evidence Rating
Relevance
high Priority