Singular value decomposition based feature extraction technique for physiological signal analysis.

Area of Science:

• Biomedical Engineering
• Signal Processing
• Computational Biology

Background:

• Multiscale entropy (MSE) is widely used to assess physiological signal complexity.
• MSE is susceptible to noise and trends, potentially leading to inaccurate complexity estimations.
• Accurate physiological signal analysis is crucial for diagnosing various health conditions.

Purpose of the Study:

• To introduce Singular Value Decomposition (SVD) as a more robust method for physiological signal feature extraction.
• To compare the efficacy of SVD against MSE in classifying physiological states.
• To evaluate the potential of SVD-based analysis for clinical applications, such as diagnosing congestive heart failure (CHF).

Main Methods:

• Physiological signals were analyzed using Singular Value Decomposition (SVD) for feature extraction.
• Support Vector Machine (SVM) was employed to classify different physiological states based on extracted features.
• A dataset from PhysioNet was utilized for performance evaluation.

Main Results:

• SVD-based feature extraction achieved a classification accuracy of 89.157%.
• This accuracy significantly surpasses the 71.084% achieved using Multiscale Entropy (MSE).
• The SVD approach demonstrated superior effectiveness in distinguishing physiological states.

Conclusions:

• Singular Value Decomposition (SVD) provides a more reliable method for analyzing physiological signal complexity compared to MSE.
• The proposed SVD and SVM analysis procedure is effective for classifying physiological states.
• This method shows potential as a diagnostic aid for conditions like congestive heart failure (CHF).