Noise Reduction of Lung Sounds based on Singular Spectrum Analysis combined with Discrete Cosine Transform

• Published in: Applied acoustics Vol. 199; p. 109005
• Main Authors: Abbasi Baharanchi, Shahrzad, Vali, Mansour, Modaresi, Mohammadreza
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.10.2022
• Subjects: Discrete Cosine Transform (DCT); Lung sound denoising; Respiratory sounds; Safety component; Singular Spectrum Analysis (SSA); Singular Value Decomposition (SVD); Singular Spectrum Analysis (SSA); Singular Value Decomposition (SVD); Lung sound denoising; Safety component; Respiratory sounds; Discrete Cosine Transform (DCT)
• ISSN: 0003-682X; 1872-910X
• DOI: 10.1016/j.apacoust.2022.109005

•Lung sound analysis plays an essential role in diagnosing lung diseases.•Bronchovesicular and vesicular breath sounds are normal lung sounds heardposteriorly.•Noise removal is necessary for enhancing the respiratory sound signals.•Singular Spectrum Analysis (SSA) decomposes the breathing sounds.•The SSA combined with Discrete Cosine Transform can denoise the pulmonary sounds. Lung sound signals are sensitive to environmental noise. Much research has been conducted on the enhancement of pulmonary sound. This study investigated the normal lung sounds as bronchovesicular (BV) and vesicular (V) signals and proposed a novel denoising method called SSA-DCT. Using Singular Spectrum Analysis (SSA), the noise-related components were separated from respiratory information components. An algorithm was also proposed to determine the information and noise time interval, and by applying Discrete Cosine Transform (DCT), the signal energy was attenuated in the noise intervals. Moreover, the concept of safety component, which is secure against noise, was introduced. Then, an algorithm for automatic identification of BV and V signals using the safety component was presented. The error of this algorithm at SNR of 10 dB was 5%. Lung sounds were recorded from 12 healthy subjects using four channels over the posterior chest wall. The signals were recorded in an acoustic laboratory and then contaminated with additive white Gaussian noise with different levels of SNR. The respiratory signals were also recorded in a relatively quiet environment with real ambient noise and denoised by the proposed method. The proposed method was compared with Coiflet wavelet decomposition with hard SureShrink thresholding. The denoising performance of both methods was evaluated using qualitative and quantitative approaches. The SSA-DCT method (e.g., at an SNR level of 10 dB) with average segmental SNR improvements of 2.52 and 3.44 dB for the BV and V signals, respectively, is significantly superior to the wavelet analysis with average segmental SNR improvements of 0.89 and 1.53 dB for the BV and V signals, respectively.