Version 1
: Received: 13 May 2024 / Approved: 13 May 2024 / Online: 13 May 2024 (10:34:24 CEST)
How to cite:
Chambi, E. M.; Cuela, J.; Zegarra, M.; Sulla, E.; Rendulich, J. Benchmarking Time-Frequency Representations of PCG Signals for Classification Valvular Heart Diseases Using Deep Features and Machine Learning. Preprints2024, 2024050829. https://doi.org/10.20944/preprints202405.0829.v1
Chambi, E. M.; Cuela, J.; Zegarra, M.; Sulla, E.; Rendulich, J. Benchmarking Time-Frequency Representations of PCG Signals for Classification Valvular Heart Diseases Using Deep Features and Machine Learning. Preprints 2024, 2024050829. https://doi.org/10.20944/preprints202405.0829.v1
Chambi, E. M.; Cuela, J.; Zegarra, M.; Sulla, E.; Rendulich, J. Benchmarking Time-Frequency Representations of PCG Signals for Classification Valvular Heart Diseases Using Deep Features and Machine Learning. Preprints2024, 2024050829. https://doi.org/10.20944/preprints202405.0829.v1
APA Style
Chambi, E. M., Cuela, J., Zegarra, M., Sulla, E., & Rendulich, J. (2024). Benchmarking Time-Frequency Representations of PCG Signals for Classification Valvular Heart Diseases Using Deep Features and Machine Learning. Preprints. https://doi.org/10.20944/preprints202405.0829.v1
Chicago/Turabian Style
Chambi, E. M., Erasmo Sulla and Jorge Rendulich. 2024 "Benchmarking Time-Frequency Representations of PCG Signals for Classification Valvular Heart Diseases Using Deep Features and Machine Learning" Preprints. https://doi.org/10.20944/preprints202405.0829.v1
Abstract
Heart sounds and murmurprovidecrucial diagnosis information for valvular heart diseases (VHD). Phonocardiogram (PCG) combined with modern digital processing techniques, provides a complementary tool for clinicians. This article proposes a benchmark different time-frequency representations, which are spectogram, mel-spectogram and cochleagram for obtaining images, in addition to the use of two interpolation techniques to improve the quality of the images, which are Bicubic and Lanczos. Deep features are extracted from a pre-trained model called VGG16 and for feature reduction the Boruta algorithm is applied. To evaluate the models and obtain more precise results, nested cross-validation is used. The best results achieved in this study were for coclegram with 99.2% accuracy and mel-spectogram representation with the bicubic interpolation technique which reached 99.4% accuracy, both having support vector machine (SVM) as classifier 10 algorithm. Overall, this study highlights the potential of time-frequency representations of PCG 11 signals combined with modern digital processing techniques and machine learning algorithms for 12 accurate diagnosis of VHD.
Computer Science and Mathematics, Signal Processing
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.