An ambient denoising method based on multi-channel non-negative matrix factorization for wheezing detection

Abstract

In this paper, a parallel computing method is proposed to perform the background denoising and wheezing detection from a multi-channel recording captured during the auscultation process. The proposed system is based on a non-negative matrix factorization (NMF) approach and a detection strategy. Moreover, the initialization of the proposed model is based on singular value decomposition to avoid dependence on the initial values of the NMF parameters. Additionally, novel update rules to simultaneously address the multichannel denoising while preserving an orthogonal constraint to maximize source separation have been designed. The proposed system has been evaluated for the task of wheezing detection showing a significant improvement over state-of-the-art algorithms when noisy sound sources are present. Moreover, parallel and high-performance techniques have been used to speedup the execution of the proposed system, showing that it is possible to achieve fast execution times, which enables its implementation in real-world scenarios.

Noise-tolerant NMF-based parallel algorithm for respiratory rate estimation

“…Regarding the computational complexity of the preprocessing stage, it is primarily influenced by the STFT computation. Following [26], a coarsegrained strategy has been implemented for parallel STFT computation, resulting in the following complexity:…”

“…Other popular approaches use NMF for noise removal in auscultation recordings [25]. NMF has shown remarkable promise in this context, particularly in scenarios where the target sound source exists in a multi-channel setup, with the target sound present in only one of these channels [26]. This approach has gained significant traction, even in recent research focused on the auscultation of newborns [27].…”

Real-time counting of wheezing events from lung sounds using deep learning algorithms: Implications for disease prediction and early intervention

“…They tended to be developed based on the learning data collected by auscultation for a short period of 10 to 70 s and labeled by clinicians [38,39]. Much of the previous work focused on addressing methodological challenges associated with noise cancellation or reduction [40,41], detection of the breathing section, or binary classification of an individual cycle of respiration [11,22,23,42,43]. Due to a lack of adaptability for real-time, continuous long-term signals, most lung sound classification algorithms have not been widely implemented in practice, with limited applicability in self-symptom management or telemedicine [2,44].…”

A Quasi-Newton Matrix Factorization-Based Model for Recommendation

“…Liu et al (2023) design a Bi-regularized Non-negative Matrix Factorization (B-NMF) method, incorporating symmetry and graph regularization, to enhance the availability of learning representation by expanding the latent factor space. NMF even received various application in the clinical (Akçay et al, 2022;Sweeney et al, 2023), genetic (Seo et al, 2022;Wu et al, 2023), autonomous vehicle (Seo et al, 2022;Wu et al, 2023), environmental (Cao et al, 2023;Muñoz-Montoro et al, 2023) and financial (Farzadnia & Vanani, 2023) domain.…”