Benchmarking of eight recurrent neural network variants for breath phase and adventitious sound detection on a self-developed open-access lung sound database-HF_Lung_V1.

A reliable, remote, and continuous real-time respiratory sound monitor with automated respiratory sound analysis ability is urgently required in many clinical scenarios-such as in monitoring disease progression of coronavirus disease 2019-to replace conventional auscultation with a handheld stethoscope. However, a robust computerized respiratory sound analysis algorithm for breath phase detection and adventitious sound detection at the recording level has not yet been validated in practical applications. In this study, we developed a lung sound database (HF_Lung_V1) comprising 9,765 audio files of lung sounds (duration of 15 s each), 34,095 inhalation labels, 18,349 exhalation labels, 13,883 continuous adventitious sound (CAS) labels (comprising 8,457 wheeze labels, 686 stridor labels, and 4,740 rhonchus labels), and 15,606 discontinuous adventitious sound labels (all crackles). We conducted benchmark tests using long short-term memory (LSTM), gated recurrent unit (GRU), bidirectional LSTM (BiLSTM), bidirectional GRU (BiGRU), convolutional neural network (CNN)-LSTM, CNN-GRU, CNN-BiLSTM, and CNN-BiGRU models for breath phase detection and adventitious sound detection. We also conducted a performance comparison between the LSTM-based and GRU-based models, between unidirectional and bidirectional models, and between models with and without a CNN. The results revealed that these models exhibited adequate performance in lung sound analysis. The GRU-based models outperformed, in terms of F1 scores and areas under the receiver operating characteristic curves, the LSTM-based models in most of the defined tasks. Furthermore, all bidirectional models outperformed their unidirectional counterparts. Finally, the addition of a CNN improved the accuracy of lung sound analysis, especially in the CAS detection tasks.

Prevalence of respiratory viruses using polymerase chain reaction in children with wheezing, a systematic review and meta-analysis.

INTRODUCTION: Wheezing is a major problem in children, and respiratory viruses are often believed to be the causative agent. While molecular detection tools enable identification of respiratory viruses in wheezing children, it remains unclear if and how these viruses are associated with wheezing. The objective of this systematic review is to clarify the prevalence of different respiratory viruses in children with wheezing. METHODS: We performed an electronic in Pubmed and Global Index Medicus on 01 July 2019 and manual search. We performed search of studies that have detected common respiratory viruses in children ≤18 years with wheezing. We included only studies using polymerase chain reaction (PCR) assays. Study data were extracted and the quality of articles assessed. We conducted sensitivity, subgroup, publication bias, and heterogeneity analyses using a random effects model. RESULTS: The systematic review included 33 studies. Rhinovirus, with a prevalence of 35.6% (95% CI 24.6-47.3, I2 98.4%), and respiratory syncytial virus, at 31.0% (95% CI 19.9-43.3, I2 96.4%), were the most common viruses detected. The prevalence of other respiratory viruses was as follows: human bocavirus 8.1% (95% CI 5.3-11.3, I2 84.6%), human adenovirus 7.7% (95% CI 2.6-15.0, I2 91.0%), influenza virus6.5% (95% CI 2.2-12.6, I2 92.4%), human metapneumovirus5.8% (95% CI 3.4-8.8, I2 89.0%), enterovirus 4.3% (95% CI 0.1-12.9, I2 96.2%), human parainfluenza virus 3.8% (95% CI 1.5-6.9, I2 79.1%), and human coronavirus 2.2% (95% CI 0.6-4.4, I2 79.4%). CONCLUSIONS: Our results suggest that rhinovirus and respiratory syncytial virus may contribute to the etiology of wheezing in children. While the clinical implications of molecular detection of respiratory viruses remains an interesting question, this study helps to illuminate the potential of role respiratory viruses in pediatric wheezing. REVIEW REGISTRATION: PROSPERO, CRD42018115128.

A basic investigation into the optimization of cylindrical tubes used as acoustic stethoscopes for auscultation in COVID-19 diagnosis.

During the COVID-19 outbreak, the auscultation of heart and lung sounds has played an important role in the comprehensive diagnosis and real-time monitoring of confirmed cases. With clinicians wearing protective clothing in isolation wards, a potato chip tube stethoscope, which is a secure and flexible substitute for a conventional stethoscope, has been used by Chinese medical workers in the first-line treatment of COVID-19. In this study, an optimal design for this simple cylindrical stethoscope is proposed based on the fundamental theory of acoustic waveguides. Analyses of the cutoff frequency, sound power transmission coefficient, and sound wave propagation in the uniform lossless tube provide theoretical guidance for selecting the geometric parameters for this simple cylindrical stethoscope. A basic investigation into the auscultatory performances of the original tube and the optimal tube with proposed dimensions was conducted both in a semi-anechoic chamber and in a quiet laboratory. Both experimental results and front-line doctors' clinical feedback endorse the proposed theoretical optimization.

Characteristics of Pulmonary Auscultation in Patients with 2019 Novel Coronavirus in China.

BACKGROUND: Effective auscultations are often hard to implement in isolation wards. To date, little is known about the characteristics of pulmonary auscultation in novel coronavirus (COVID-19) pneumonia. OBJECTIVES: The aim of this study was to explore the features and clinical significance of pulmonary auscultation in COVID-19 pneumonia using an electronic stethoscope in isolation wards. METHODS: This cross-sectional, observational study was conducted among patients with laboratory-confirmed COVID-19 at Wuhan Red-Cross Hospital during the period from January 27, 2020, to February 12, 2020. Standard auscultation with an electronic stethoscope was performed and electronic recordings of breath sounds were analyzed. RESULTS: Fifty-seven patients with average age of 60.6 years were enrolled. The most common symptoms were cough (73.7%) during auscultation. Most cases had bilateral lesions (96.4%) such as multiple ground-glass opacities (69.1%) and fibrous stripes (21.8%). High-quality auscultation recordings (98.8%) were obtained, and coarse breath sounds, wheezes, coarse crackles, fine crackles, and Velcro crackles were identified. Most cases had normal breath sounds in upper lungs, but the proportions of abnormal breath sounds increased in the basal fields where Velcro crackles were more commonly identified at the posterior chest. The presence of fine and coarse crackles detected 33/39 patients with ground-glass opacities (sensitivity 84.6% and specificity 12.5%) and 8/9 patients with consolidation (sensitivity 88.9% and specificity 15.2%), while the presence of Velcro crackles identified 16/39 patients with ground-glass opacities (sensitivity 41% and specificity 81.3%). CONCLUSIONS: The abnormal breath sounds in COVID-19 pneumonia had some consistent distributive characteristics and to some extent correlated with the radiologic features. Such evidence suggests that electronic auscultation is useful to aid diagnosis and timely management of the disease. Further studies are indicated to validate the accuracy and potential clinical benefit of auscultation in detecting pulmonary abnormalities in COVID-19 infection.

Time series and fractal analyses of wheezing: a novel approach.

Since the outbreak of the pandemic Coronavirus Disease 2019, the world is in search of novel non-invasive methods for safer and early detection of lung diseases. The pulmonary pathological symptoms reflected through the lung sound opens a possibility of detection through auscultation and of employing spectral, fractal, nonlinear time series and principal component analyses. Thirty-five signals of vesicular and expiratory wheezing breath sound, subjected to spectral analyses shows a clear distinction in terms of time duration, intensity, and the number of frequency components. An investigation of the dynamics of air molecules during respiration using phase portrait, Lyapunov exponent, sample entropy, fractal dimension, and Hurst exponent helps in understanding the degree of complexity arising due to the presence of mucus secretions and constrictions in the respiratory airways. The feature extraction of the power spectral density data and the application of principal component analysis helps in distinguishing vesicular and expiratory wheezing and thereby, giving a ray of hope in accomplishing an early detection of pulmonary diseases through sound signal analysis.

A Wearable Stethoscope for Long-Term Ambulatory Respiratory Health Monitoring.

Lung sounds acquired by stethoscopes are extensively used in diagnosing and differentiating respiratory diseases. Although an extensive know-how has been built to interpret these sounds and identify diseases associated with certain patterns, its effective use is limited to individual experience of practitioners. This user-dependency manifests itself as a factor impeding the digital transformation of this valuable diagnostic tool, which can improve patient outcomes by continuous long-term respiratory monitoring under real-life conditions. Particularly patients suffering from respiratory diseases with progressive nature, such as chronic obstructive pulmonary diseases, are expected to benefit from long-term monitoring. Recently, the COVID-19 pandemic has also shown the lack of respiratory monitoring systems which are ready to deploy in operational conditions while requiring minimal patient education. To address particularly the latter subject, in this article, we present a sound acquisition module which can be integrated into a dedicated garment; thus, minimizing the role of the patient for positioning the stethoscope and applying the appropriate pressure. We have implemented a diaphragm-less acousto-electric transducer by stacking a silicone rubber and a piezoelectric film to capture thoracic sounds with minimum attenuation. Furthermore, we benchmarked our device with an electronic stethoscope widely used in clinical practice to quantify its performance.

A 14-year Prospective Study of Human Coronavirus Infections in Hospitalized Children: Comparison With Other Respiratory Viruses.

BACKGROUND: Human coronaviruses (HCoVs) have been recognized as causative agents of respiratory tract infections.Our aim was to describe HCoV infections in hospitalized children in a prospective surveillance study for 14 years and compare them with other respiratory viruses. METHODS: As a part of an ongoing prospective study to identify the etiology of viral respiratory infections in Spain, we performed the analysis of HCoV infections in children hospitalized in a secondary hospital in Madrid, between October 2005 and June 2018. Clinical data of HCoV patients were compared with those infected by rhinovirus, respiratory syncytial virus and influenza. RESULTS: The study population consisted of 5131 hospitalizations for respiratory causes in children. A total of 3901 cases (75.9%) had a positive viral identification and 205 cases (4.1%) were positive for HCoV. Only 41 cases (20%) of HCoV infection were detected as single infections. Episodes of recurrent wheezing were the most common diagnosis, and 112 children (54%) had hypoxia. Clinical data in HCoV cases were similar to those associated with rhinovirus; however, patients with HCoV were younger. Other viruses were associated with hypoxia more frequently than cases with HCoV; high fever was more common in influenza infections and bronchiolitis in respiratory syncytial virus group. Although a slight peak of circulation appears mostly in winter, HCoV has been detected throughout the year as well. CONCLUSIONS: HCoV infections represent a small fraction of respiratory infections that require hospitalization in children and their characteristics do not differ greatly from other respiratory viral infections.