Evaluation of the usefulness of spectral analysis of inspiratory lung sounds recorded with phonopneumography in patients with interstitial pneumonia

Deep learning diagnostic and severity-stratification for interstitial lung diseases and chronic obstructive pulmonary disease in digital lung auscultations and ultrasonography: clinical protocol for an observational case-control study

BACKGROUND

Interstitial lung diseases (ILD), such as idiopathic pulmonary fibrosis (IPF) and non-specific interstitial pneumonia (NSIP), and chronic obstructive pulmonary disease (COPD) are severe, progressive pulmonary disorders with a poor prognosis. Prompt and accurate diagnosis is important to enable patients to receive appropriate care at the earliest possible stage to delay disease progression and prolong survival. Artificial intelligence-assisted lung auscultation and ultrasound (LUS) could constitute an alternative to conventional, subjective, operator-related methods for the accurate and earlier diagnosis of these diseases. This protocol describes the standardised collection of digitally-acquired lung sounds and LUS images of adult outpatients with IPF, NSIP or COPD and a deep learning diagnostic and severity-stratification approach.

METHODS

A total of 120 consecutive patients (≥ 18 years) meeting international criteria for IPF, NSIP or COPD and 40 age-matched controls will be recruited in a Swiss pulmonology outpatient clinic, starting from August 2022. At inclusion, demographic and clinical data will be collected. Lung auscultation will be recorded with a digital stethoscope at 10 thoracic sites in each patient and LUS images using a standard point-of-care device will be acquired at the same sites. A deep learning algorithm (DeepBreath) using convolutional neural networks, long short-term memory models, and transformer architectures will be trained on these audio recordings and LUS images to derive an automated diagnostic tool. The primary outcome is the diagnosis of ILD versus control subjects or COPD. Secondary outcomes are the clinical, functional and radiological characteristics of IPF, NSIP and COPD diagnosis. Quality of life will be measured with dedicated questionnaires. Based on previous work to distinguish normal and pathological lung sounds, we estimate to achieve convergence with an area under the receiver operating characteristic curve of > 80% using 40 patients in each category, yielding a sample size calculation of 80 ILD (40 IPF, 40 NSIP), 40 COPD, and 40 controls.

DISCUSSION

This approach has a broad potential to better guide care management by exploring the synergistic value of several point-of-care-tests for the automated detection and differential diagnosis of ILD and COPD and to estimate severity. Trial registration Registration: August 8, 2022.

CLINICALTRIALS

gov Identifier: NCT05318599.

The coming era of a new auscultation system for analyzing respiratory sounds

Auscultation with stethoscope has been an essential tool for diagnosing the patients with respiratory disease. Although auscultation is non-invasive, rapid, and inexpensive, it has intrinsic limitations such as inter-listener variability and subjectivity, and the examination must be performed face-to-face. Conventional stethoscope could not record the respiratory sounds, so it was impossible to share the sounds. Recent innovative digital stethoscopes have overcome the limitations and enabled clinicians to store and share the sounds for education and discussion. In particular, the recordable stethoscope made it possible to analyze breathing sounds using artificial intelligence, especially based on neural network. Deep learning-based analysis with an automatic feature extractor and convoluted neural network classifier has been applied for the accurate analysis of respiratory sounds. In addition, the current advances in battery technology, embedded processors with low power consumption, and integrated sensors make possible the development of wearable and wireless stethoscopes, which can help to examine patients living in areas of a shortage of doctors or those who need isolation. There are still challenges to overcome, such as the analysis of complex and mixed respiratory sounds and noise filtering, but continuous research and technological development will facilitate the transition to a new era of a wearable and smart stethoscope.

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.

A machine-learning based approach to quantify fine crackles in the diagnosis of interstitial pneumonia: A proof-of-concept study

Fine crackles are frequently heard in patients with interstitial lung diseases (ILDs) and are known as the sensitive indicator for ILDs, although the objective method for analyzing respiratory sounds including fine crackles is not clinically available. We have previously developed a machine-learning-based algorithm which can promptly analyze and quantify the respiratory sounds including fine crackles. In the present proof-of-concept study, we assessed the usefulness of fine crackles quantified by this algorithm in the diagnosis of ILDs.We evaluated the fine crackles quantitative values (FCQVs) in 60 participants who underwent high-resolution computed tomography (HRCT) and chest X-ray in our hospital. Right and left lung fields were evaluated separately.In sixty-seven lung fields with ILDs in HRCT, the mean FCQVs (0.121 ± 0.090) were significantly higher than those in the lung fields without ILDs (0.032 ± 0.023, P < .001). Among those with ILDs in HRCT, the mean FCQVs were significantly higher in those with idiopathic pulmonary fibrosis than in those with other types of ILDs (P = .002). In addition, the increased mean FCQV was associated with the presence of traction bronchiectasis (P = .003) and honeycombing (P = .004) in HRCT. Furthermore, in discriminating ILDs in HRCT, an FCQV-based determination of the presence or absence of fine crackles indicated a higher sensitivity compared to a chest X-ray-based determination of the presence or absence of ILDs.We herein report that the machine-learning-based quantification of fine crackles can predict the HRCT findings of lung fibrosis and can support the prompt and sensitive diagnosis of ILDs.

The acoustic characteristics of fine crackles predict honeycombing on high-resolution computed tomography

Background

Honeycombing on high-resolution computed tomography (HRCT) is a distinguishing feature of usual interstitial pneumonia and predictive of poor outcome in interstitial lung diseases (ILDs). Although fine crackles are common in ILD patients, the relationship between their acoustic features and honeycombing on HRCT has not been well characterized.

Methods

Lung sounds were digitally recorded from 71 patients with fine crackles and ILD findings on chest HRCT. Lung sounds were analyzed by fast Fourier analysis using a sound spectrometer (Easy-LSA; Fukuoka, Japan). The relationships between the acoustic features of fine crackles in inspiration phases (onset timing, number, frequency parameters, and time-expanded waveform parameters) and honeycombing in HRCT were investigated using multivariate logistic regression analysis.

Results

On analysis, the presence of honeycombing on HRCT was independently associated with onset timing (early vs. not early period; odds ratios [OR] 10.407, 95% confidence interval [95% CI] 1.366–79.298, P = 0.024), F99 value (the percentile frequency below which 99% of the total signal power is accumulated) (unit Hz = 100; OR 5.953, 95% CI 1.221–28.317, P = 0.029), and number of fine crackles in the inspiratory phase (unit number = 5; OR 4.256, 95% CI 1.098–16.507, P = 0.036). In the receiver-operating characteristic curves for number of crackles and F99 value, the cutoff levels for predicting the presence of honeycombing on HRCT were calculated as 13.2 (area under the curve [AUC], 0.913; sensitivity, 95.8%; specificity, 75.6%) and 752 Hz (AUC, 0.911; sensitivity, 91.7%; specificity, 85.2%), respectively. The multivariate logistic regression analysis additionally using these cutoff values revealed an independent association of number of fine crackles in the inspiratory phase, F99 value, and onset timing with the presence of honeycombing (OR 33.907, 95% CI 2.576–446.337, P = 0.007; OR 19.397, 95% CI 2.311–162.813, P = 0.006; and OR 12.383, 95% CI 1.443–106.293, P = 0.022; respectively).

Conclusions

The acoustic properties of fine crackles distinguish the honeycombing from the non-honeycombing group. Furthermore, onset timing, number of crackles in the inspiratory phase, and F99 value of fine crackles were independently associated with the presence of honeycombing on HRCT. Thus, auscultation routinely performed in clinical settings combined with a respiratory sound analysis may be predictive of the presence of honeycombing on HRCT.

A novel feature extraction technique for pulmonary sound analysis based on EMD

BACKGROUND AND OBJECTIVE

The stethoscope based auscultation technique is a primary diagnostic tool for chest sound analysis. However, the performance of this method is limited due to its dependency on physicians experience, knowledge and also clarity of the signal. To overcome this problem we need an automated computer-aided diagnostic system that will be competent in noisy environment. In this paper, a novel feature extraction technique is introduced for discriminating various pulmonary dysfunctions in an automated way based on pattern recognition algorithms.

METHOD

In this work, the disease correlated relevant characteristics of lung sounds signals are identified in terms of statistical distribution parameters: mean, variance, skewness, and kurtosis. These features are extracted from selective morphological components of the mapped signal in the empirical mode decomposition domain. The feature set is fed to the classifier model to differentiate their corresponding classes.

RESULTS

The significance of features developed are validated by conducting several experiments using supervised and unsupervised classifiers. Furthermore, the discriminating power of the proposed features is compared with three types of baseline features. The experimental result is evaluated by statistical analysis and also validated with physicians inference.

CONCLUSIONS

It is found that the proposed features extraction technique is superior to the baseline methods in terms of classification accuracy, sensitivity and specificity. The developed method gives better results compared to baseline methods in any circumstance. The proposed method gives a higher accuracy of 94.16, sensitivity of 100 and specificity of 93.75 for an artificial neural network classifier.

Linear and Nonlinear Analysis of Base Lung Sound in Extrinsic Allergic Alveolitis Patients in Comparison to Healthy Subjects

Objective: Pulmonary disorders are frequently characterized by the presence of adventitious sounds added to the breathing or base lung sound (BLS). The aim of this work was to assess the features of BLS in extrinsic allergic alveolitis (EAA) patients in comparison to healthy subjects, applying linear and nonlinear analysis techniques.

Methods: We investigated the multichannel lung sounds on the posterior chest of 16 females, 8 healthy and 8 EAA patients, when breathing at 1.5 L/s. BLS linear features were obtained from the power spectral density (PSD) while nonlinear features were extracted by the concepts of irregularity and complexity, i.e., spectral, sample and multi-scale entropy.

Results: The results demonstrated that spectral percentiles of BLS were lower in EAA patients than in healthy subjects but statistical significance (p<0.05) was obtained only for expiration at the left apical and both basal regions. Also, the maximum amplitude of the PSD in patients reached statistical significance ( p < 0.05) for the expiratory phase at basal regions. In the case of nonlinear techniques, significant lower values ( p < 0.05) were obtained for EAA patients during both respiratory phases at left apical and both basal regions.

Conclusion: In conclusion, we found that BLS in chronic EAA patients is characterized by lower spectral percentiles, lower irregularity and lower complexity than in healthy subjects suggesting the feasibility of its clinical usefulness by screening its temporal alteration.

Respiratory sounds classification using statistical biomarker

In this paper, we have proposed a new feature extraction technique based on statistical morphology of lung sound signal (LS). This work attempts to (i) generate certain intrinsic mode functions (IMFs), (ii) select a set of informative IMFs and (iii) extract relevant features from the selected IMFs and residue. Feature vector is formed by using the higher order moments: mean, standard deviation, skewness and kurtosis and employed as input to the classifier models for classification of three types of LS signals: crackle, wheeze and normal. The efficiency of these features is examined with an artificial neural network (ANN) classifier and compared the results with three baseline methods. The proposed method gives a superior performance in term of classification accuracy, sensitivity and specificity.

Recent advances in connective tissue disease related interstitial lung disease

INTRODUCTION

Interstitial lung disease (ILD) is a common manifestation of connective tissue disease (CTD). Although the majority of patients with CTD-ILD are stable or slowly progressive, a significant group exhibits a more severe and progressive decline. Interstitial pneumonia with autoimmune features (IPAF) describes the subset of patients with interstitial pneumonia who have features suggesting underlying autoimmunity, but whose features fall short of a clear diagnosis of CTD. Areas covered: In this focused review, we discuss recent advances in early detection, prognostic evaluation, and management of autoimmune forms of ILD. Expert commentary: Early detection of ILD and a better understanding of factors that impact prognostication may be helpful when making decisions regarding therapeutic interventions. The treatment of CTD-ILD should be comprehensive, is often fraught with challenges and can be complicated by comorbid conditions and extra-thoracic disease activities. Several large randomized studies have examined the impact of immunosuppressive therapy for CTD-ILD, however, additional studies are needed to determine the optimal treatment strategies. Future studies may provide additional information about the best treatments in patients with IPAF.

Automatic adventitious respiratory sound analysis: A systematic review

Background

Automatic detection or classification of adventitious sounds is useful to assist physicians in diagnosing or monitoring diseases such as asthma, Chronic Obstructive Pulmonary Disease (COPD), and pneumonia. While computerised respiratory sound analysis, specifically for the detection or classification of adventitious sounds, has recently been the focus of an increasing number of studies, a standardised approach and comparison has not been well established.

Objective

To provide a review of existing algorithms for the detection or classification of adventitious respiratory sounds. This systematic review provides a complete summary of methods used in the literature to give a baseline for future works.

Data sources

A systematic review of English articles published between 1938 and 2016, searched using the Scopus (1938-2016) and IEEExplore (1984-2016) databases. Additional articles were further obtained by references listed in the articles found. Search terms included adventitious sound detection, adventitious sound classification, abnormal respiratory sound detection, abnormal respiratory sound classification, wheeze detection, wheeze classification, crackle detection, crackle classification, rhonchi detection, rhonchi classification, stridor detection, stridor classification, pleural rub detection, pleural rub classification, squawk detection, and squawk classification.

Study selection

Only articles were included that focused on adventitious sound detection or classification, based on respiratory sounds, with performance reported and sufficient information provided to be approximately repeated.

Data extraction

Investigators extracted data about the adventitious sound type analysed, approach and level of analysis, instrumentation or data source, location of sensor, amount of data obtained, data management, features, methods, and performance achieved.

Data synthesis

A total of 77 reports from the literature were included in this review. 55 (71.43%) of the studies focused on wheeze, 40 (51.95%) on crackle, 9 (11.69%) on stridor, 9 (11.69%) on rhonchi, and 18 (23.38%) on other sounds such as pleural rub, squawk, as well as the pathology. Instrumentation used to collect data included microphones, stethoscopes, and accelerometers. Several references obtained data from online repositories or book audio CD companions. Detection or classification methods used varied from empirically determined thresholds to more complex machine learning techniques. Performance reported in the surveyed works were converted to accuracy measures for data synthesis.

Limitations

Direct comparison of the performance of surveyed works cannot be performed as the input data used by each was different. A standard validation method has not been established, resulting in different works using different methods and performance measure definitions.

Conclusion

A review of the literature was performed to summarise different analysis approaches, features, and methods used for the analysis. The performance of recent studies showed a high agreement with conventional non-automatic identification. This suggests that automated adventitious sound detection or classification is a promising solution to overcome the limitations of conventional auscultation and to assist in the monitoring of relevant diseases.

Novel algorithm to identify and differentiate specific digital signature of breath sound in patients with diffuse parenchymal lung disease

BACKGROUND AND OBJECTIVE

Auscultation is an important part of the clinical examination of different lung diseases. Objective analysis of lung sounds based on underlying characteristics and its subsequent automatic interpretations may help a clinical practice.

METHODS

We collected the breath sounds from 8 normal subjects and 20 diffuse parenchymal lung disease (DPLD) patients using a newly developed instrument and then filtered off the heart sounds using a novel technology. The collected sounds were thereafter analysed digitally on several characteristics as dynamical complexity, texture information and regularity index to find and define their unique digital signatures for differentiating normality and abnormality. For convenience of testing, these characteristic signatures of normal and DPLD lung sounds were transformed into coloured visual representations. The predictive power of these images has been validated by six independent observers that include three physicians.

RESULTS

The proposed method gives a classification accuracy of 100% for composite features for both the normal as well as lung sound signals from DPLD patients. When tested by independent observers on the visually transformed images, the positive predictive value to diagnose the normality and DPLD remained 100%.

CONCLUSIONS

The lung sounds from the normal and DPLD subjects could be differentiated and expressed according to their digital signatures. On visual transformation to coloured images, they retain 100% predictive power. This technique may assist physicians to diagnose DPLD from visual images bearing the digital signature of the condition.

A comparative analysis of machine learning methods for classification type decision problems in healthcare

Advanced analytical techniques are gaining popularity in addressing complex classification type decision problems in many fields including healthcare and medicine. In this exemplary study, using digitized signal data, we developed predictive models employing three machine learning methods to diagnose an asthma patient based solely on the sounds acquired from the chest of the patient in a clinical laboratory. Although, the performances varied slightly, ensemble models (i.e., Random Forest and AdaBoost combined with Random Forest) achieved about 90% accuracy on predicting asthma patients, compared to artificial neural networks models that achieved about 80% predictive accuracy. Our results show that non-invasive, computerized lung sound analysis that rely on low-cost microphones and an embedded real-time microprocessor system would help physicians to make faster and better diagnostic decisions, especially in situations where x-ray and CT-scans are not reachable or not available. This study is a testament to the improving capabilities of analytic techniques in support of better decision making, especially in situations constraint by limited resources.

Computerized lung sound analysis as diagnostic aid for the detection of abnormal lung sounds: a systematic review and meta-analysis

RATIONALE

The standardized use of a stethoscope for chest auscultation in clinical research is limited by its inherent inter-listener variability. Electronic auscultation and automated classification of recorded lung sounds may help prevent some of these shortcomings.

OBJECTIVE

We sought to perform a systematic review and meta-analysis of studies implementing computerized lung sound analysis (CLSA) to aid in the detection of abnormal lung sounds for specific respiratory disorders.

METHODS

We searched for articles on CLSA in MEDLINE, EMBASE, Cochrane Library and ISI Web of Knowledge through July 31, 2010. Following qualitative review, we conducted a meta-analysis to estimate the sensitivity and specificity of CLSA for the detection of abnormal lung sounds.

MEASUREMENTS AND MAIN RESULTS

Of 208 articles identified, we selected eight studies for review. Most studies employed either electret microphones or piezoelectric sensors for auscultation, and Fourier Transform and Neural Network algorithms for analysis and automated classification of lung sounds. Overall sensitivity for the detection of wheezes or crackles using CLSA was 80% (95% CI 72-86%) and specificity was 85% (95% CI 78-91%).

CONCLUSIONS

While quality data on CLSA are relatively limited, analysis of existing information suggests that CLSA can provide a relatively high specificity for detecting abnormal lung sounds such as crackles and wheezes. Further research and product development could promote the value of CLSA in research studies or its diagnostic utility in clinical settings.