RESUMO
Heart sound signals are vital for the machine-assisted detection of congenital heart disease. However, the performance of diagnostic results is limited by noise during heart sound acquisition. A limitation of existing noise reduction schemes is that the pathological components of the signal are weak, which have the potential to be filtered out with the noise. In this research, a novel approach for classifying heart sounds based on median ensemble empirical mode decomposition (MEEMD), Hurst analysis, improved threshold denoising, and neural networks are presented. In decomposing the heart sound signal into several intrinsic mode functions (IMFs), mode mixing and mode splitting can be effectively suppressed by MEEMD. Hurst analysis is adopted for identifying the noisy content of IMFs. Then, the noise-dominated IMFs are denoised by an improved threshold function. Finally, the noise reduction signal is generated by reconstructing the processed components and the other components. A database of 5000 heart sounds from congenital heart disease and normal volunteers was constructed. The Mel spectral coefficients of the denoised signals were used as input vectors to the convolutional neural network for classification to verify the effectiveness of the preprocessing algorithm. An accuracy of 93.8%, a specificity of 93.1%, and a sensitivity of 94.6% were achieved for classifying the normal cases from abnormal one.
RESUMO
Aiming at the problems of obscure clinical auscultation features of pulmonary hypertension associated with congenital heart disease and the complexity of existing machine-aided diagnostic algorithms, an algorithm based on the statistical characteristics of the high-frequency components of the second heart sound signal is proposed. Firstly, an endpoint detection adaptive segmentation method is employed to extract the second heart sounds. Subsequently, the high-frequency component of the heart sound is decomposed using the discrete wavelet transform. Statistical features including the Hurst exponent, Lempel-Ziv information and sample entropy are extracted from this component. Finally, the extracted features are utilized to train an extreme gradient boosting algorithm (XGBoost) classifier, which achieves an accuracy of 80.45% in triple classification. Notably, this method eliminates the need for a noise reduction algorithm, allows for swift feature extraction, and achieves effective multi-classification using only three features. It is promising for early screening of pulmonary hypertension associated with congenital heart disease.
Assuntos
Cardiopatias Congênitas , Ruídos Cardíacos , Hipertensão Pulmonar , Humanos , Processamento de Sinais Assistido por Computador , Hipertensão Pulmonar/diagnóstico , Algoritmos , Cardiopatias Congênitas/complicações , Cardiopatias Congênitas/diagnósticoRESUMO
Introduction: Congenital heart disease (CHD) is a cardiovascular disorder caused by structural defects in the heart. Early screening holds significant importance for the effective treatment of this condition. Heart sound analysis is commonly employed to assist in the diagnosis of CHD. However, there is currently a lack of an efficient automated model for heart sound classification, which could potentially replace the manual process of auscultation. Methods: This study introduces an innovative and efficient screening and classification model, combining a locally concatenated fusion approach with a convolutional neural network based on coordinate attention (LCACNN). In this model, Mel-frequency spectral coefficients (MFSC) and envelope features are locally fused and employed as input to the LCACNN network. This model automatically analyzes feature map energy information, eliminating the need for denoising processes. Discussion: The proposed classification model in this study demonstrates a robust capability for identifying congenital heart disease, potentially substituting manual auscultation to facilitate the detection of patients in remote areas. Results: This study introduces an innovative and efficient screening and classification model, combining a locally concatenated fusion approach with a convolutional neural network based on coordinate attention (LCACNN). In this model, Mel-frequency spectral coefficients (MFSC) and envelope features are locally fused and employed as input to the LCACNN network. This model automatically analyzes feature map energy information, eliminating the need for denoising processes. To assess the performance of the classification model, comparative ablation experiments were conducted, achieving classification accuracies of 91.78% and 94.79% on the PhysioNet and HS databases, respectively. These results significantly outperformed alternative classification models.