Key result
A convolutional neural network using PCG gammatonegrams accurately detects heart abnormalities.
Why the study?
Does a gammatonegram-based CNN model accurately classify PCG signals into healthy and unhealthy categories compared to existing methods?
Does a gammatonegram-based CNN model accurately classify PCG signals into healthy and unhealthy categories compared to existing methods?
A CNN model using gammatonegram images of PCG signals can accurately detect heart abnormalities.
Phonocardiogram (PCG) consists of heart sound signals and carries the important information associated with the functioning of the heart. In this work, a method is presented for the detection of heart disease based on gammatonegram images of PCG signals. Initially, PCG signals are converted into time-frequency spectrum images using gammatonegram. The obtained gammatonegram images are resized and augmented. The augmented images are fed to Convolutional Neural Network (CNN). The CNN model classifies the PCG signals into healthy and unhealthy categories. The performance of the model is measured with a publicly available dataset. To compare the performance, Adam, Adagrad, Adadelta, RMSprop, and SGD optimizers are used. The performance of the proposed method is also compared with the existing state of art methods. On comparison, it is found that the gammatonegram images of PCG signals exhibit important information, and can be used to detect heart abnormalities with significant accuracy.
No takes yet. Share an insight, caveat, or question.
Satyasai et al. (2023) studied Heart disease. Gammatonegram based CNN model vs. Existing state of art methods was evaluated on Classification accuracy of healthy vs unhealthy PCG signals. A Convolutional Neural Network model using gammatonegram images of phonocardiogram signals can detect heart abnormalities with significant accuracy.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: