The role of respiratory function tests in infants with stridor: diagnosis at glance and follow-up.

BACKGROUND: Recently, the development of advanced, noninvasive methods has allowed the study of respiratory function even in uncooperative infants. To date, there is still little data on the application of this technique in infants with suspected airway obstruction. THE AIMS OF OUR STUDY WERE: - To evaluate the role of respiratory function testing (PFR) in the diagnosis and follow-up of infants with stridor - To evaluate the differences between patients with inspiratory stridor and expiratory stridor. - To evaluate the concordance between PFR and endoscopy. METHODS: We enrolled infants aged < 1 year with a diagnosis of inspiratory and/or expiratory chronic stridor and a group of healthy controls. For each patient we performed PFR at diagnosis (T0) and for cases at follow-up, at 3 months (T1), 6 months (T2), 12 months (T3). At T0, all patients were classified according to a clinical score, and at follow-up, stature-ponderal growth was assessed. When clinically indicated, patients underwent bronchoscopy. RESULTS: We enrolled 48 cases (42 diagnosed with inspiratory stridor and 6 expiratory stridor) and 26 healthy controls. At T0, patients with stridor had increased inspiratory time (p < 0.0001) and expiratory time (p < 0.001) than healthy controls and abnormal curve morphology depending on the type of stridor. At T0, patients with expiratory stridor had a reduced Peak expiratory flow (p < 0.023) and a longer expiratory time (p < 0.004) than patients with inspiratory stridor. We showed an excellent concordance between PFR and endoscopic examination (k = 0.885, p < 0.0001). At follow-up, we showed a progressive increase of the respiratory parameters in line with the growth. CONCLUSIONS: PFR could help improve the management of these patients through rapid and noninvasive diagnosis, careful monitoring, and early detection of those most at risk.

Automated Interpretation of Lung Sounds by Deep Learning in Children With Asthma: Scoping Review and Strengths, Weaknesses, Opportunities, and Threats Analysis.

BACKGROUND: The interpretation of lung sounds plays a crucial role in the appropriate diagnosis and management of pediatric asthma. Applying artificial intelligence (AI) to this task has the potential to better standardize assessment and may even improve its predictive potential. OBJECTIVE: This study aims to objectively review the literature on AI-assisted lung auscultation for pediatric asthma and provide a balanced assessment of its strengths, weaknesses, opportunities, and threats. METHODS: A scoping review on AI-assisted lung sound analysis in children with asthma was conducted across 4 major scientific databases (PubMed, MEDLINE Ovid, Embase, and Web of Science), supplemented by a gray literature search on Google Scholar, to identify relevant studies published from January 1, 2000, until May 23, 2023. The search strategy incorporated a combination of keywords related to AI, pulmonary auscultation, children, and asthma. The quality of eligible studies was assessed using the ChAMAI (Checklist for the Assessment of Medical Artificial Intelligence). RESULTS: The search identified 7 relevant studies out of 82 (9%) to be included through an academic literature search, while 11 of 250 (4.4%) studies from the gray literature search were considered but not included in the subsequent review and quality assessment. All had poor to medium ChAMAI scores, mostly due to the absence of external validation. Identified strengths were improved predictive accuracy of AI to allow for prompt and early diagnosis, personalized management strategies, and remote monitoring capabilities. Weaknesses were the heterogeneity between studies and the lack of standardization in data collection and interpretation. Opportunities were the potential of coordinated surveillance, growing data sets, and new ways of collaboratively learning from distributed data. Threats were both generic for the field of medical AI (loss of interpretability) but also specific to the use case, as clinicians might lose the skill of auscultation. CONCLUSIONS: To achieve the opportunities of automated lung auscultation, there is a need to address weaknesses and threats with large-scale coordinated data collection in globally representative populations and leveraging new approaches to collaborative learning.

The analysis of lung sounds in infants and children with a history of wheezing/asthma using an automatic procedure.

BACKGROUND: Lung sound analysis parameters have been reported to be useful biomarkers for evaluating airway condition. We developed an automatic lung sound analysis software program for infants and children based on lung sound spectral curves of frequency and power by leveraging machine learning (ML) technology. METHODS: To put this software program into clinical practice, in Study 1, the reliability and reproducibility of the software program using data from younger children were examined. In Study 2, the relationship between lung sound parameters and respiratory flow (L/s) was evaluated using data from older children. In Study 3, we conducted a survey using the ATS-DLD questionnaire to evaluate the clinical usefulness. The survey focused on the history of wheezing and allergies, among healthy 3-year-old infants, and then measured lung sounds. The clinical usefulness was evaluated by comparing the questionnaire results with the results of the new lung sound parameters. RESULTS: In Studies 1 and 2, the parameters of the new software program demonstrated excellent reproducibility and reliability, and were not affected by airflow (L/s). In Study 3, infants with a history of wheezing showed lower FAP0 and RPF75p (p < 0.001 and p = 0.025, respectively) and higher PAP0 (p = 0.001) than healthy infants. Furthermore, infants with asthma/asthma-like bronchitis showed lower FAP0 (p = 0.002) and higher PAP0 (p = 0.001) than healthy infants. CONCLUSIONS: Lung sound parameters obtained using the ML algorithm were able to accurately assess the respiratory condition of infants. These parameters are useful for the early detection and intervention of childhood asthma.

Neural network analysis of pharyngeal sounds can detect obstructive upper respiratory disease in brachycephalic dogs.

Brachycephalic obstructive airway syndrome (BOAS) is a highly prevalent respiratory disease affecting popular short-faced dog breeds such as Pugs and French bulldogs. BOAS causes significant morbidity, leading to poor exercise tolerance, sleep disorders and a shortened lifespan. Despite its severity, the disease is commonly missed by owners or disregarded by veterinary practitioners. A key clinical sign of BOAS is stertor, a low-frequency snoring sound. In recent years, a functional grading scheme has been introduced to semi-objectively grade BOAS based on the presence of stertor and other abnormal signs. However, correctly grading stertor requires significant experience and adding an objective component would aid accuracy and repeatability. This study proposes a recurrent neural network model to automatically detect and grade stertor in laryngeal electronic stethoscope recordings. The model is developed using a novel dataset of 665 labelled recordings taken from 341 dogs with diverse BOAS clinical signs. Evaluated via nested cross validation, the neural network predicts the presence of clinically significant BOAS with an area under the receiving operating characteristic of 0.85, an operating sensitivity of 71% and a specificity of 86%. The algorithm could enable widespread screening for BOAS to be conducted by both owners and veterinarians, improving treatment and breeding decisions.

ConvLSNet: A lightweight architecture based on ConvLSTM model for the classification of pulmonary conditions using multichannel lung sound recordings.

Characterization of lung sounds (LS) is indispensable for diagnosing respiratory pathology. Although conventional neural networks (NNs) have been widely employed for the automatic diagnosis of lung sounds, deep neural networks can potentially be more useful than conventional NNs by allowing accurate classification without requiring preprocessing and feature extraction. Utilizing the long short-term memory (LSTM) layers to reveal the sequence-based properties of the LS time series, a novel architecture consisting of a cascade of convolutional long short-term memory (ConvLSTM) and LSTM layers, namely ConvLSNet is developed, which permits highly accurate diagnosis of pulmonary disease states. By modeling the multichannel lung sounds through the ConvLSTM layer, the proposed ConvLSNet architecture can concurrently deal with the spatial and temporal properties of the six-channel LS recordings without heavy preprocessing or data transformation. Notably, the proposed model achieves a classification accuracy of 97.4 % based on LS data corresponding to three pulmonary conditions, namely asthma, COPD, and the healthy state. Compared with architectures consisting exclusively of CNN or LSTM layers, as well as those employing a cascade integration of 2DCNN and LSTM layers, the proposed ConvLSNet architecture exhibited the highest classification accuracy, while imposing the lowest computational cost as quantified by the number of parameters, training time, and learning rate.

Association of Cough Severity with Asthma Control and Quality of Life in Patients with Severe Asthma.

PURPOSE: Symptoms are important components in determining asthma control and in the adjustment of treatment levels. However, clinical relevance of cough in severe asthma is not well-understood. This study aimed to evaluate the severity and association of cough with patient-reported outcomes (PROs) in patients with severe asthma. METHODS: This study analyzed cross-sectional data from the Korean Severe Asthma Registry. The severity of coughing and wheezing symptoms was assessed using a Visual Analog Scale (VAS) ranging from 0 to 100 for each symptom. Additionally, PROs included the Asthma Control Test (ACT), the Severe Asthma Questionnaire (SAQ), and the EuroQoL 5-Dimension (EQ-5D) index. Multivariate linear regression analysis was employed to explore the relationship between cough severity and other PRO scores. RESULTS: A total of 498 patients with severe asthma (age: 57.9 ± 13.1 years, females: 60.2%) were analyzed. The cough VAS score was higher than the wheeze score (median 30, [interquartile range 10-50] vs. 20 [0-50]; P < 0.001). Additionally, 22.5% of patients ranked in a higher tertile for cough severity compared to wheezing, while 18.5% ranked higher for wheezing severity than cough. Significant correlations were observed between cough and wheeze VAS scores (r = 0.61, P < 0.05) and between each symptom's VAS score and the SAQ (cough: r = -0.41, P < 0.001; wheeze: r = -0.52, P < 0.001), ACT scores (cough: r = -0.50, P < 0.001; wheeze: r = -0.63, P < 0.001) and EQ-5D index (cough: r = -0.40, P < 0.001; wheeze: r = -0.45, P < 0.001). In univariate regression analysis, the cough VAS score had weaker descriptive power (R2) values than the wheeze VAS score in relation to the PRO measures. Nevertheless, cough severity remained significantly associated with ACT, SAQ scores and EQ-5D index in multivariate analyses adjusted for wheeze severity and other confounders. CONCLUSION: Cough frequently presents as a severe symptom in patients with severe asthma and could have distinct impact on asthma control and quality of life.

Severity of sleep apnea as a prognostic factor for mortality in patients with multiple system atrophy.

BACKGROUND: We determined whether the severity of sleep apnea increases the risk of mortality in patients with multiple system atrophy (MSA) with and without stridor. MethodsThis retrospective study included patients who underwent polysomnography within one year after diagnosis of probable MSA. Stridor, sleep apnea, and arousal from sleep were determined using full-night polysomnography. Disease severity was measured using the Unified MSA Rating Scale (UMSARS). Survival data were collected and analyzed using Cox regression analysis. RESULTS: Sixty-four patients with MSA were included. During a median follow-up of 34.5 months, 49 (76.6 %) patients died. Stridor was present in 56.3 % of patients. Patients with stridor had more severe sleep apnea and shorter sleep time than those without, but the hazard ratio (HR) for death did not differ between patients with and without stridor. Among patients without stridor, apnea-hypopnea index ≥30/h (HR, 6.850; 95 % confidence interval [CI], 1.983-23.664; p = 0.002) and a score of UMSARS I + II (HR, 1.080; 95 % CI, 1.040-1.121; p < 0.001) were independently associated with death. In contrast, among patients with stridor, frequent arousals from sleep (HR, 0.254; 95 % CI, 0.089-0.729; p = 0.011) were a significant factor associated with longer survival, while MSA-cerebellar type tended to be associated with poor survival (HR, 2.195; 95 % CI, 0.941-5.120; p = 0.069). CONCLUSION: The severity of sleep apnea might be a significant predictor of shorter survival in MSA patients without stridor, whereas frequent arousals from sleep might be a significant predictor for longer survival in MSA patients with stridor.

Relationships between lung function, allergy, and wheezing in urban children.

BACKGROUND: Allergic sensitization and low lung function in early childhood are risk factors for subsequent wheezing and asthma. However, it is unclear how allergic sensitization affects lung function over time. OBJECTIVE: We sought to test whether allergy influences lung function and whether these factors synergistically increase the risk of continued wheezing in childhood. METHODS: We analyzed longitudinal measurements of lung function (spirometry and impulse oscillometry) and allergic sensitization (aeroallergen skin tests and serum allergen-specific IgE) throughout early childhood in the Urban Environmental and Childhood Asthma study, which included high-risk urban children living in disadvantaged neighborhoods. Intraclass correlation coefficients were calculated to assess lung function stability. Cluster analysis identified low, medium, and high allergy trajectories, which were compared with lung function and wheezing episodes in linear regression models. A variable selection model assessed predictors at age 5 years for continued wheezing through age 12 years. RESULTS: Lung function adjusted for growth was stable (intraclass correlation coefficient, 0.5-0.7) from age 5 to 12 years and unrelated to allergy trajectory. Lung function and allergic sensitization were associated with wheezing episodes in an additive fashion. In children with asthma, measuring lung function at age 5 years added little to the medical history for predicting future wheezing episodes through age 12 years. CONCLUSIONS: In high-risk urban children, age-related trajectories of allergic sensitization were not associated with lung function development; however, both indicators were related to continued wheezing. These results underscore the importance of understanding early-life factors that negatively affect lung development and suggest that treating allergic sensitization may not alter lung function development in early to mid-childhood.

Longitudinal tracking of intrabreath respiratory impedance in preschool children.

BACKGROUND: Longitudinal measurements of intrabreath respiratory impedance (Zrs) in preschool-aged children may be able to distinguish abnormal lung function trajectories in children with a history of wheezing compared to healthy ones. METHODS: Children from a prospective, longitudinal community-based cohort performed annual intrabreath oscillometry (IB-OSC) measurements from age 3- to 7-years. IB-OSC was performed using a single 10 Hz sinusoid while clinically asymptomatic. Linear mixed-effects models were developed to explore the effects of wheezing phenotypes, growth, and sex on seven IB-OSC outcome variables over time: resistance at end-expiration (ReE), resistance at end-inspiration (ReI), the tidal change in resistance (∆R=ReE-ReI), reactance at end-expiration (XeE), reactance at end-inspiration (XeI), the tidal change in reactance (∆X=XeE-XeI), and ∆X normalized by tidal volume (∆X/VT). RESULTS: Eighty-five children produced 374 acceptable IB-OSC measurements. Subjects were classified into one of three wheeze groups: never (n = 36), transient (n = 34), or persistent (n = 15). After adjusting for height, children with persistent wheezing, compared to those who never wheezed, had +0.814 hPa s L-1 ReE (95% confidence interval [CI] +0.178 to +1.451, p = 0.015), -0.792 hPa s L-1 XeE (95% CI -1.203 to -0.381, p = 0.003), -0.538 hPa s L-1 ∆X (95% CI -0.834 to -0.242, p = 0.007) and -1.672 hPa s L-2 ∆X/VT (95% CI -2.567 to -0.777, p < 0.001). Increasing height had a significant effect on all IB-OSC resistance and reactance variables when adjusted for the effect of preschool wheezing. CONCLUSIONS: IB-OSC is feasible for tracking lung function growth in preschool-aged children and may allow abnormal lung function to be identified early in asymptomatic preschoolers with a history of persistent wheezing.

Clinical implications of airway obstruction with normal or low FEV1 in childhood and adolescence.

BACKGROUND: Airway obstruction is defined by spirometry as a low forced expiratory volume in 1 s (FEV1) to forced vital capacity (FVC) ratio. This impaired ratio may originate from a low FEV1 (classic) or a normal FEV1 in combination with a large FVC (dysanaptic). The clinical implications of dysanaptic obstruction during childhood and adolescence in the general population remain unclear. AIMS: To investigate the association between airway obstruction with a low or normal FEV1 in childhood and adolescence, and asthma, wheezing and bronchial hyperresponsiveness (BHR). METHODS: In the BAMSE (Barn/Child, Allergy, Milieu, Stockholm, Epidemiology; Sweden) and PIAMA (Prevention and Incidence of Asthma and Mite Allergy; the Netherlands) birth cohorts, obstruction (FEV1:FVC ratio less than the lower limit of normal, LLN) at ages 8, 12 (PIAMA only) or 16 years was classified as classic (FEV1

Respiratory Diseases Diagnosis Using Audio Analysis and Artificial Intelligence: A Systematic Review.

Respiratory diseases represent a significant global burden, necessitating efficient diagnostic methods for timely intervention. Digital biomarkers based on audio, acoustics, and sound from the upper and lower respiratory system, as well as the voice, have emerged as valuable indicators of respiratory functionality. Recent advancements in machine learning (ML) algorithms offer promising avenues for the identification and diagnosis of respiratory diseases through the analysis and processing of such audio-based biomarkers. An ever-increasing number of studies employ ML techniques to extract meaningful information from audio biomarkers. Beyond disease identification, these studies explore diverse aspects such as the recognition of cough sounds amidst environmental noise, the analysis of respiratory sounds to detect respiratory symptoms like wheezes and crackles, as well as the analysis of the voice/speech for the evaluation of human voice abnormalities. To provide a more in-depth analysis, this review examines 75 relevant audio analysis studies across three distinct areas of concern based on respiratory diseases' symptoms: (a) cough detection, (b) lower respiratory symptoms identification, and (c) diagnostics from the voice and speech. Furthermore, publicly available datasets commonly utilized in this domain are presented. It is observed that research trends are influenced by the pandemic, with a surge in studies on COVID-19 diagnosis, mobile data acquisition, and remote diagnosis systems.

A 21-Year-Old Man With Dyspnea, Wheezing, and Cough.

CASE PRESENTATION: A 21-year-old male college student presented for a second opinion with low alpha-1 antitrypsin (AAT) levels and complaints of episodic dyspnea with wheezing and cough. He was a never smoker with a medical history of frequent respiratory tract infections in early childhood and allergy to dander, dust mites, peanuts, and eggs. There was no travel history outside of the continental United States. His mother had asthma. His symptoms were not controlled on inhaled corticosteroids and bronchodilators. His AAT genotype was found to be PI∗SZ, and augmentation therapy (with pooled human-plasma derived AAT) was recommended locally.

Vedolizumab-associated pulmonary manifestations in children with ulcerative colitis.

WHAT IS KNOWN AND OBJECTIVE: The off-label use of vedolizumab (VDZ) for inflammatory bowel disease in children is increasing. We report on possibly the first case of VDZ-associated pulmonary manifestations in paediatrics. CASE SUMMARY: This report details the case of a 13-year-old child with ulcerative colitis who was initiated on VDZ due to persistent active disease. After the first three doses, he developed a persistent and productive cough. Microbiological work-up was normal. VDZ discontinuation led to the resolution of symptoms. WHAT IS NEW AND CONCLUSION: To our knowledge, this is the first case report of VDZ-associated pulmonary manifestations in paediatrics. A direct, pro-inflammatory effect of VDZ has been hypothesized, but further studies are warranted.

Are children born by cesarean delivery at higher risk for respiratory sequelae?

BACKGROUND: Globally, the number of children born by cesarean delivery is constantly increasing. However, hormonal and physiological changes associated with labor and vaginal delivery are considered necessary for lung maturation. OBJECTIVE: We aimed to assess whether the mode of delivery is associated with changes in respiratory and atopic outcomes during infancy and at school age. STUDY DESIGN: We included 578 children, born at ≥37 weeks of gestation, from a prospective birth cohort study. We compared weekly respiratory symptoms throughout the first year of life and infant lung function (tidal breathing and multiple-breath washout) at 5 weeks of age between children born by cesarean delivery (N=114) and those born by vaginal delivery (N=464) after term pregnancy in healthy women. At a follow-up visit conducted at 6 years of age (N=371, of which 65 were delivered by cesarean delivery), we assessed respiratory, atopic, and lung function outcomes (spirometry, body plethysmography, and multiple-breath washout). We performed adjusted regression analyses to examine the association between cesarean delivery and respiratory and atopic outcomes. To account for multiple testing, we used the Bonferroni correction, which led to an adapted significance level of P<.002. RESULTS: During infancy, children born by cesarean delivery did not have more respiratory symptoms than those born by vaginal delivery (median, 4 weeks; interquartile range, 7 weeks vs median, 5 weeks; interquartile range, 7 weeks; adjusted incidence rate ratio, 0.8; 95% confidence interval, 0.6-1.0; P=.02). Infant lung function was similar between the groups. Children born by cesarean delivery did not have a higher incidence of "ever wheezing" (adjusted odds ratio, 0.9; 95% confidence interval, 0.5-1.8; P=.78) or current asthma (adjusted odds ratio, 0.4; 95% confidence interval, 0.0-3.5; P=.42) at school age than those born by vaginal delivery. There was no difference in the lung function parameters between the groups. CONCLUSION: Cesarean delivery was not associated with respiratory symptoms in the first year of life, nor with different respiratory or atopic outcomes at school age, when compared with vaginal delivery. Our results indicate that there are no long-term consequences on the respiratory health of the child associated with cesarean delivery.

Respiratory sound classification for crackles, wheezes, and rhonchi in the clinical field using deep learning.

Auscultation has been essential part of the physical examination; this is non-invasive, real-time, and very informative. Detection of abnormal respiratory sounds with a stethoscope is important in diagnosing respiratory diseases and providing first aid. However, accurate interpretation of respiratory sounds requires clinician's considerable expertise, so trainees such as interns and residents sometimes misidentify respiratory sounds. To overcome such limitations, we tried to develop an automated classification of breath sounds. We utilized deep learning convolutional neural network (CNN) to categorize 1918 respiratory sounds (normal, crackles, wheezes, rhonchi) recorded in the clinical setting. We developed the predictive model for respiratory sound classification combining pretrained image feature extractor of series, respiratory sound, and CNN classifier. It detected abnormal sounds with an accuracy of 86.5% and the area under the ROC curve (AUC) of 0.93. It further classified abnormal lung sounds into crackles, wheezes, or rhonchi with an overall accuracy of 85.7% and a mean AUC of 0.92. On the other hand, as a result of respiratory sound classification by different groups showed varying degree in terms of accuracy; the overall accuracies were 60.3% for medical students, 53.4% for interns, 68.8% for residents, and 80.1% for fellows. Our deep learning-based classification would be able to complement the inaccuracies of clinicians' auscultation, and it may aid in the rapid diagnosis and appropriate treatment of respiratory diseases.

Atopic dermatitis trajectories to age 8 years in the GUSTO cohort.

BACKGROUND: The heterogeneity of childhood atopic dermatitis (AD) underscores the need to understand latent phenotypes that may inform risk stratification and disease prognostication. OBJECTIVE: To identify AD trajectories across the first 8 years of life and investigate risk factors associated with each trajectory and their relationships with other comorbidities. METHODS: Data were collected prospectively from 1152 mother-offspring dyads in the Growing Up in Singapore Towards healthy Outcomes (GUSTO) cohort from ages 3 months to 8 years. AD was defined based on parent-reported doctor's diagnosis. An unsupervised machine learning technique was used to determine AD trajectories. RESULTS: Three AD trajectories were identified as follows: early-onset transient (6.3%), late-onset persistent (6.3%) and early-onset persistent (2.1%), alongside a no AD/reference group (85.2%). Early-onset transient AD was positively associated with male gender, family history of atopy, house dust mite sensitization and some measures of wheezing. Early-onset persistent AD was associated with antenatal/intrapartum antibiotic use, food sensitization and some measures of wheezing. Late-onset persistent AD was associated with a family history of atopy, some measures of house dust mite sensitization and some measures of allergic rhinitis and wheezing. CONCLUSION AND CLINICAL RELEVANCE: Three AD trajectories were identified in this birth cohort, with different risk factors and prognostic implications. Further work is needed to understand the molecular and immunological origins of these phenotypes.

Follow-up study of airway microbiota in children with persistent wheezing.

BACKGROUND: Increasing evidence revealed that airway microbial dysbiosis was associated with increased risk of asthma, or persistent wheezing (PW). However, the role of lung microbiota in PW or wheezing recurrence remains poorly understood. METHODS: In this prospective observational study, we performed a longitudinal 16S rRNA-based microbiome survey on bronchoalveolar lavage (BAL) samples collected from 35 infants with PW and 28 age-matched infants (control group). A 2-year follow-up study on these PW patients was conducted. The compositions of lower airway microbiota were analyzed at the phylum and genus levels. RESULTS: Our study showed a clear difference in lower airway microbiota between PW children and the control group. Children with PW had a higher abundance of Elizabethkingia and Rothia, and lower abundance of Fusobacterium compared with the control group. At the end of the 2-year follow-up, 20 children with PW (57.1%) experienced at least one episode of wheezing, and 15 (42.9%) did not suffer from wheezing episodes. Furthermore, PW children with recurrence also had increased abundances of Elizabethkingia and Rothia relative to those who had no recurrence. Additionally, wheezing history, different gender, and caesarean section demonstrated a greater impact in airway microbiota compositions. CONCLUSION: This study suggests that the alterations of lower airway microbiota could be strongly associated with the development of wheezing, and early airway microbial changes could also be associated with wheezing recurrence later in life.

Magnetic resonance imaging with gradient sound respiration guide.

Respiratory motion management is crucial for high-resolution MRI of the heart, lung, liver and kidney. In this article, respiration guide using acoustic sound generated by pulsed gradient waveforms was introduced in the pulmonary ultrashort echo time (UTE) sequence and validated by comparing with retrospective respiratory gating techniques. The validated sound-guided respiration was implemented in non-contrast enhanced renal angiography. In the sound-guided respiration, breathe-in and-out instruction sounds were generated with sinusoidal gradient waveforms with two different frequencies (602 and 321 Hz). Performance of the sound-guided respiration was evaluated by measuring sharpness of the lung-liver interface with a 10-90% rise distance, w10-90, and compared with three respiratory motion managements in a free-breathing UTE scan: without respiratory gating (w/o gating), 0-dimensional k-space navigator (k-point navigator), and image-based self-gating (Img-SG). The sound-guided respiration was implemented in stack-of-stars balanced steady-state free precession with inversion recovery preparation for renal angiography. No subjects reported any discomfort or inconvenience with the sound-guided respiration in pulmonary or renal MRI scans. The lung-liver interface of the UTE images for sound-guided respiration (w10-90 = 6.99 ± 2.90 mm), k-point navigator (8.51 ± 2.71 mm), and Img-SG (7.01 ± 2.06 mm) was significantly sharper than that for w/o gating (17.13 ± 2.91 mm; p < 0.0001 for all of sound-guided respiration, k-point navigator and Img-SG). Sharpness of the lung-liver interface was comparable between sound-guided respiration and Img-SG (p = 0.99), but sound-guided respiration achieved better visualization of pulmonary vasculature. Renal angiography with the sound-guided respiration clearly delineated renal, segmental and interlobar arteries. In conclusion, the gradient sound guided respiration can facilitate a consistent diaphragm position in every breath and achieve performance of respiratory motion management comparable to image-based self-gating.

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.