Pulmonary Auscultation Research Articles

TÉCNICAS FISIOTERAPÊUTICAS NOS DIFERENTES NÍVEIS DE GRAVIDADE DA BRONQUIOLITE

Acute viral bronchiolitis (AVB) is the main respiratory infection in children under two years, responsible for the highest number of hospital admissions in this age group in the winter period. It presents different levels of severity, with treatment varying according to the severity of each frame, however, there is still no consensus on the performance of physiotherapy. Objective: To identify and relate the physiotherapeutic techniques used in the different levels of AVB severity. Method: A prospective, observational, cross-sectional study conducted in a public hospital of medium complexity in the metropolitan region of the south of the country. Neonatal characteristics of the child and the physiotherapeutic techniques performed were associated with the severity level of respiratory dysfunction. Results: The sample consisted of 31 infants, with a higher prevalence of males (64.5%) and moderate severity of AVB (51.4%), according to the instrument adopted for evaluation (Wood-Downes scale modified by Ferrés). Increased expiratory flow, associated with thoracic compression and decompression maneuvers, were the most used physiotherapeutic techniques (35.5%). There was no statistically significant difference in the association between bronchiolitis conduct and severity, possibly because the sample did not include serious cases. However, the relationship between initial pulmonary auscultation and techniques used (p=0.04) was observed. Conclusion: The choice of the physiotherapeutic techniques adopted was based on the findings of the initial pulmonary auscultation and not only on the characteristics of the disease. Despite consistency in intervention, as advocated in international consensus and recommendations, conventional techniques without scientific evidence were also used.

Clinical analysis of influenza in the neonatal intensive care unit

BackgroundThe aim was to investigate the clinical characteristics, treatment and prognosis of neonatal influenza.MethodsThe clinical data of 21 neonates who were diagnosed with influenza and admitted to the neonatal intensive care unit of Henan Provincial Children’s Hospital, China, between January 2023 and January 2024 were retrospectively analyzed.ResultsA total of 21 patients were admitted, including 14 with influenza A and 7 with influenza B. Eighteen of these patients were reported to have been exposed to family members with respiratory symptoms before hospitalization. Among all the patients’ mothers, only 1 received the influenza vaccine during pregnancy. Fifteen newborns had fever, 13 appetite loss, 10 cough, 9 shortness of breath, 9 nasal obstruction, 3 runny nose, 3 vomiting, 2 severe wheezing, 2 choking, 2 diarrhea, 1 bloating, and 1 sputum in the throat. The pulmonary auscultation sounds were coarse in 19 neonates, weak in 2, moist rales were appreciated in 5 and wheezing in 4 of them. The peripheral total white blood cell count was normal in 18 patients and elevated in 3. The C-reactive protein level was normal in all subjects, and the procalcitonin level was elevated in 1. Nineteen patients had pneumonia on chest imaging. All patients were treated with oseltamivir and finally recovered.ConclusionInfluenza A is the most common type of neonatal influenza. The clinical symptoms are atypical, and fever is the main symptom. Treatment with oseltamivir is safe and effective, and the prognosis is mostly favorable.

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

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. 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. 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). 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. 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.

Detection of Mycoplasma pneumoniae in hospitalized children with pneumonia in Laos

Mycoplasma pneumoniae has been described worldwide as an important cause of community-acquired pneumonia. From December 2013 to December 2014, 461 children admitted to Mahosot Hospital, Vientiane, Laos, with acute respiratory infection were investigated for upper respiratory microorganisms using probe-based real-time polymerase chain reaction (PCR) (FTD33). M. pneumoniae was detected by FTD33 in the upper respiratory tract of three patients, two girls and one boy, 5.7 and 3.9 years old and 13.6 years old, respectively. They presented with clinical features compatible with M. pneumoniae infection. The two girls were also positive for other potential pathogens. The boy had abnormal pulmonary auscultation, and one of the girls had significant anaemia. These results suggest that enhancement of diagnostic systems for M. pneumoniae detection and analysis of its antibiotic resistance profile is needed to raise awareness and improve understanding of the epidemiology of M. pneumoniae infection in Laos, enable targeted therapy, and inform treatment guidelines.

FOREIGN BODY ASPIRATION IN CHILDREN: CLINICAL, RADIOLOGICAL AND BRONCHOSCOPY FINDINGS

Background: Foreign body aspiration (FBA) is a serious problem that leads to partial or complete airway obstruction. Respiratory distress and pneumonia are the most common complications. The assortment of foreign bodies is very wide. The most common include candy, fish bones, peanuts and nuts, toys, food and batteries. Aim of this study was to determine how accurate the presenting symptoms and signs of foreign body aspiration are and to assess the radiologic findings, the types and sites of the foreign bodies removed. Patients and Methods: a retrospective study included101 patient from age of one month to 15 years with suspected foreign body aspiration based on history, clinical examination, Chest Xray. Rigid bronchoscopy was done for patients under general anaesthesia with muscle relaxation using neuromuscular blocking agents. For every patient gender, age, residence, interval between event and symptom onset, symptoms at admission, signs of respiratory insufficiency, pulmonary auscultation findings, Chest X-ray findings, interval between admission and bronchoscopy, the location of foreign body, type of foreign body, hospitalization days were studied and statistically analysed. A p-value of < 0.05 was considered statistically significant. Results: Seventy six (75.24%) were proved to have foreign body aspiration that was removed by rigid bronchoscopy. toddler was the common age group 43(42.6%). Male were affected more than females 1.2:1. A significantly higher percentage of patients who had a foreign body were witnessed by a family member than those who did not have a foreign body 61.8% vs 0% (p=0.001). Stridor and cough with breathlessness were the most common signs and symptoms while unilateral wheezes, crepitations and diminished air entry were less frequent while each of crepitations, unilateral wheezes and cough with breathlessness were significantly associated with foreign body aspiration (p=0.001, 0.001, 0.02 respectively). Abnormal chest X ray finding was significantly associated with FBA (p=0.02) with hyperinflation being the most frequent finding. The types of foreign bodies removed were sunflower seeds (27.6%), food particles (18.4%) and nuts (17.1%). The most common site of foreign body was the right main bronchus (52.6%) followed by left main bronchus (34.2%). There is a significant association between foreign body aspiration and the time interval between admission and bronchoscopy but no significant relation with age, interval between the event and onset of symptoms and hospitalization days. Conclusion: The history of being witnessed by a family member, the presence of stridor and cough and finding of unilateral wheezes and crepitations on examination as well as hyperinflation on chest X ray are significantly associated with FBA. The sunflower seeds and food particles are the most common types and the right main bronchus is the main site of foreign bodies removed by bronchoscopy.

Abuso de Substância Ilícitas, Sibilos e Redução do Rendimento Desportivo Numa Atleta de MMA: Haverá Alguma Relação?

This fighter's case report addresses the importance of anamnesis and physical examination. The initial consultation was requested due to a simple complaint of arm allergy. During the clinical interview it was found a drop on physical performance, a history of illicit drugs abuse and wheezing on pulmonary auscultation. With this information it was possible to reach a surprising diagnosis, which is rare in adults and could have had a fatal outcome. The treatment was carried out successfully and the athlete returned to high-performance sports without restrictions.

“A double twist” presentation – a case report of purulent cardiac tamponade following a rare complication of small-cell lung cancer radiotherapy

Background: Small cell lung cancer is an aggressive tumor with a poor prognosis that requires prompt treatment. While radiotherapy may enhance survival when superior vena cava syndrome is present, radiation therapy–induced pericardial disease can be a potential complication. Case Report: A 55-year-old man, who recently underwent radiotherapy for stage IV small-cell lung cancer complicated by superior vena cava syndrome, presented with chest pain and dyspnea. In the emergency room, he was dyspneic, hypotensive, and tachycardic. Pulmonary auscultation revealed the absence of lung sounds on the right. The initial electrocardiogram showed ST-segment elevation in lateral leads and in lead DII, with reciprocal changes in lead DIII. A bedside transthoracic echocardiogram revealed cardiac tamponade and emergent pericardiocentesis was performed, removing 500 ml of purulent fluid, resulting in an immediate clinical improvement. Thoracentesis was also performed, showing no empyema. Large spectrum empirical antibiotic therapy was started. Cultures from the pericardial fluid and peripheral blood grew multi-sensitive Streptococcus pneumoniae. Cytological analysis of the pericardial fluid was consistent with infection. The patient improved after 2 weeks of targeted antibiotic therapy and underwent the first cycle of chemotherapy. He was discharged with an early scheduled pulmonology appointment. Conclusions: Although the most common causes of pericardial effusion in lung cancer are malignant, non-malignant etiologies should also be considered. This patient had an infectious pericardial effusion most probably due to a pericardial-mediastinal mass fistula caused by radiotherapy. This was a diagnostic challenge, both in the emergency room as well in the inpatient setting.

Automatic breathing phase identification based on the second derivative of the recorded lung sounds

This paper presents a new method for automatic identification of the inspiratory and expiratory breathing phases in lung sound recordings. Adventitious lung sounds (wheezes and crackles), superimposed on the breath sounds, are generally an early indication of the disease, and their timing in the breathing cycle (early/mid/late inspiratory or expiratory) has clinical significance for monitoring or diagnosing disease. Therefore, the identification of the phases of the breathing cycle is an essential step for clinical interpretation of pulmonary auscultation. The proposed algorithm is designed to be robust in the presence of adventitious lung sounds or where the breath sounds may be noisy compared to healthy lung sounds. The algorithm uses the Savitzky & Golay (SG) filter to estimate the second derivative of the lung sound signal then calculates its normalized absolute value. A threshold value is used to clip any large amplitude peaks and, following low-pas filtering, the breathing phases are visible in the plotted signal. A rule-based approach to locating peaks and troughs is then used to identify inspirations and expirations. The performance of the method is evaluated using four different datasets: (a) a longitudinal dataset recorded from 19 subjects with a diagnosis of idiopathic pulmonary fibrosis (IPF), (b) a cross-sectional dataset recorded from 55 subjects who were referred for high-resolution computed tomography (HRCT) scan of the chest for various clinical indications, (c) a longitudinal dataset recorded from 10 healthy subjects, and (d) an open access lung sounds dataset containing recordings from 41 subjects with wheeze (9 with chronic obstructive pulmonary fibrosis and 32 with asthma). On average for inspiratory phase identification the algorithm had a sensitivity of (mean (standard deviation)) 92.84 (12.88)%, positive predictive value of 90.64(15.96)%, and F1-score of 90.67 (12.53)%. For identification of the expiratory phase, the algorithm had an average sensitivity of 92.19 (13.44)%, an average positive predictive value of 91.55 (15.23)%, and an average F1-score of 90.89 (12.45)%. The method shows good potential for automatic identification of breathing phases in recorded lung sounds.

Artificial intelligence-based framework to identify the abnormalities in the COVID-19 disease and other common respiratory diseases from digital stethoscope data using deep CNN.

The utilization of lung sounds to diagnose lung diseases using respiratory sound features has significantly increased in the past few years. The Digital Stethoscope data has been examined extensively by medical researchers and technical scientists to diagnose the symptoms of respiratory diseases. Artificial intelligence-based approaches are applied in the real universe to distinguish respiratory disease signs from human pulmonary auscultation sounds. The Deep CNN model is implemented with combined multi-feature channels (Modified MFCC, Log Mel, and Soft Mel) to obtain the sound parameters from lung-based Digital Stethoscope data. The model analysis is observed with max-pooling and without max-pool operations using multi-feature channels on respiratory digital stethoscope data. In addition, COVID-19 sound data and enriched data, which are recently acquired data to enhance model performance using a combination of L2 regularization to overcome the risk of overfitting because of less respiratory sound data, are included in the work. The suggested DCNN with Max-Pooling on the improved dataset demonstrates cutting-edge performance employing a multi-feature channels spectrogram. The model has been developed with different convolutional filter sizes (, , , , and ) that helped to test the proposed neural network. According to the experimental findings, the suggested DCNN architecture with a max-pooling function performs better to identify respiratory disease symptoms than DCNN without max-pooling. In order to demonstrate the model's effectiveness in categorization, it is trained and tested with the DCNN model that extract several modalities of respiratory sound data.

Bronquiolitis obliterante posinfecciosa secundaria a adenovirus

Bronchiolitis obliterans is a rare chronic obstructive respiratory disease secondary to damage of the lower respiratory tract causing inflammation and partial or total obliteration of it. There are different causes, being infectious the most frequent in pediatrics, mainly due to adenovirus. We present the case of an 18-month-old male infant with a history of admission to the Pediatric Intensive Care Unit at 8 months of age due to bronchitis secondary to respiratory syncytial virus and adenovirus. After this episode, he presented persistent respiratory distress and pathological pulmonary auscultation. High resolution pulmonary computed tomography showed a mosaic pattern with areas of air trapping and decreased vascular caliber in the affected areas, findings suggestive of bronchiolitis obliterans.

Risk factors and effect of dyspnea inappropriate treatment in adults' emergency department: a retrospective cohort study.

Dyspnea is a frequent symptom in adults' emergency departments (EDs). Misdiagnosis at initial clinical examination is common, leading to early inappropriate treatment and increased in-hospital mortality. Risk factors of inappropriate treatment assessable at early examination remain undescribed herein. The objective of this study was to identify clinical risk factors of dyspnea and inappropriate treatment in patients admitted to ED. This is an observational retrospective cohort study. Patients over the age of 15 who were admitted to adult EDs of the University Hospital of Toulouse (France) with dyspnea were included from 1 July to 31 December 2019. The primary end-point was dyspnea and inappropriate treatment was initiated at ED. Inappropriate treatment was defined by looking at the final diagnosis of dyspnea at hospital discharge and early treatment provided. Afterward, this early treatment at ED was compared to the recommended treatment defined by the International Guidelines for Acute Heart Failure, bacterial pneumonia, chronic obstructive pulmonary disease, asthma or pulmonary embolism. A total of 2123 patients were analyzed. Of these, 809 (38%) had inappropriate treatment in ED. Independent risk factors of inappropriate treatment were: age over 75 years (OR, 1.46; 95% CI, 1.18-1.81), history of heart disease (OR, 1.32; 95% CI, 1.07-1.62) and lung disease (OR, 1.47; 95% CI, 1.21-1.78), SpO 2 <90% (OR, 1.64; 95% CI, 1.37-2.02), bilateral rale (OR, 1.25; 95% CI, 1.01-1.66), focal cracklings (OR, 1.32; 95% CI, 1.05-1.66) and wheezing (OR, 1.62; 95% CI, 1.31-2.03). In multivariate analysis, under-treatment significantly increased in-hospital mortality (OR, 2.13; 95% CI, 1.29-3.52) compared to appropriate treatment. Over-treatment nonsignificantly increased in-hospital mortality (OR, 1.43; 95% CI, 0.99-2.06). Inappropriate treatment is frequent in patients admitted to ED for dyspnea. Patients older than 75 years, with comorbidities (heart or lung disease), hypoxemia (SpO 2 <90%) or abnormal pulmonary auscultation (especially wheezing) are at risk of inappropriate treatment.

Clinical decision support system using a machine learning model to assist simultaneous cardiopulmonary auscultation: Open-label randomized controlled trial.

The utility of a clinical decision support system using a machine learning (ML) model for simultaneous cardiac and pulmonary auscultation is unknown. This study aimed to develop and evaluate an ML system's utility for cardiopulmonary auscultation. First, we developed an ML system for cardiopulmonary auscultation, using cardiopulmonary sound files from our previous study. The technique involved pre-processing, feature extraction, and classification through several neural network layers. After integration, the output class was categorized as "normal," "abnormal," or "undetermined." Second, we evaluated the ML system with 24 junior residents in an open-label randomized controlled trial at a university hospital. Participants were randomly assigned to the ML system group (intervention) or conventional auscultation group (control). During training, participants listened to four cardiac and four pulmonary sounds, all of which were correctly classified. Then, participants classified a series of 16 simultaneous cardiopulmonary sounds. The control group auscultated the sounds using noise-cancelling headphones, while the intervention group did so by watching recommendations from the ML system. The total scores for correctly identified normal or abnormal cardiopulmonary sounds in the intervention group were significantly higher than those in the control group (366/384 [95.3%] vs. 343/384 [89.3%], P = 0.003). The cardiac test score in the intervention group was better (111/192 [57.8%] vs. 90/192 [46.9%], P = 0.04); there was no significant difference in pulmonary auscultation. The ML-based system improved the accuracy of cardiopulmonary auscultation for junior residents. This result suggests that the system can assist early-career physicians in accurate screening.

Usefulness of Lung Ultrasound as a Method for Early Diagnosis of Interstitial Lung Disease in Patients with Rheumatoid Arthritis.

To determine the value of lung ultrasound (LUS) compared to high-resolution computed tomography (HRCT) in the early diagnosis of interstitial lung disease (ILD) in patients with rheumatoid arthritis (RA). An observational prospective study was performed. Were included patients with respiratory symptoms or/and, patients with crackles in auscultation during medical consultation. All patients underwent to chest X-rays, LUS, HRCT,and respiratory function tests. A total of 192 patients with RA were included. Mean disease duration was 16.8 ± 11.1 years. 72% were positive for rheumatoid factor or anti-citrullinated antibodies. Of the total number of subjects, 54.7% had respiratory symptoms. The other patients did not have respiratory symptoms, but they did have had crackles on pulmonary auscultation. B lines > 11.5 on the ROC curve predicted ILD (AUC 0.63; CI 95%: 0.55-0.71; p < 0.003). A DLCO value of <7.13 significantly predicted the presence of ILD (AUC 0.61; 95% CI: 0.52-0.70; p < 0.028). The findings of this study suggest that LUS is a valuable tool for the early diagnosis of ILD in patients with RA, and together with DLCO, can adequately predict the presence of ILD in this population. LUS also helps to determine which patients with respiratory symptoms and signs suggestive for ILD are undergo to HRCT.

Pneumonia, aerossaculite e osteomilelite por Aspergillus fumigatus em um papagaio-cinzento (Psittacus erithacus) - relato de caso

ABSTRACT Aspergillosis is a fungal disease with high morbidity and mortality in wild and exotic bird species. The aim of this report is to describe a case of acute aspergillosis caused by Aspergillus fumigatus in a young African grey parrot (Psittacus erithacus). A three-month-old male African grey parrot from a commercial breeding site presented serous nasal discharge, inappetence, and wheezing on pulmonary auscultation. The parrot died three days after the onset of clinical signs. Postmortem evaluation demonstrated multiple smooth gray plaques in air sacs and left lung with dark red areas and adhesions to the ribs. Microscopically, there were intralesional hyphae and conidiophores in the lungs, air sacs, and ribs, which were associated with pneumonia, airsacculitis, and osteomyelitis, respectively. DNA samples were extracted from paraffinized tissues and subjected to PCR targeting the ITS-2 region, followed by sequencing, which yielded a sequence with 100% coverage and 100% identity to Aspergillus fumigatus sequences. Although A. fumigatus infection is quite common in birds, a particular aspect of interest in this case was the finding of conidiophores in the bone marrow, which may occur in birds due to air circulation through pneumatic bones.

Pitiose esofágica e gástrica em cão

ABSTRACT: Pythiosis is the disease caused by aquatic oomycetes of the genus Pythium. In dogs the typical course of the disease involves the gastrointestinal tract, characterized by transmural thickening of the stomach or intestines. However, infection by Pythium spp. has only rarely been recognized as a cause of esophagitis in dogs. Thus, the present reports a case of esophageal and gastric pythiosis in an eight-month-old female pit bull dog. The dog was attended at the hospital after two months presenting regurgitation and dyspnea. It was reported that the dog lived in an urban area and had no previous sanitary issues. At clinical examination it was noted that the dog presented crackling sounds at pulmonary auscultation. A support therapy accompanied by antibiotics has been employed; however, 22 days after hospitalization the clinical condition worsened, and the dog died. At necropsy, the wall of the distal segment of the esophagus and the cardia and part of the fundus of the stomach were expanded by a focal extensive irregular intramural annular mass. Additionally, there was a transmural esophageal fistula. At histology, the walls of the esophagus and stomach were extensively expanded by multifocal extensive areas of necrosis, associated with a pyogranulomatous infiltrate and abundant granulation tissue containing multiple negative images of hyphae that were highlighted by silver impregnation (Grocott). Furthermore, immunohistochemistry and PCR for P. insidiosum were both positive in samples of paraffin-embedded esophageal tissue.

Multimodal Integration of Enhanced Novel Pulmonary Auscultation Real-time Diagnostic System

Multimodal Integration of Enhanced Novel Pulmonary Auscultation Real-time Diagnostic System

Enabling Continuous Breathing-Phase Contextualization via Wearable-Based Impedance Pneumography and Lung Sounds: A Feasibility Study.

Chronic respiratory diseases affect millions and are leading causes of death in the US and worldwide. Pulmonary auscultation provides clinicians with critical respiratory health information through the study of Lung Sounds (LS) and the context of the breathing-phase and chest location in which they are measured. Existing auscultation technologies, however, do not enable the simultaneous measurement of this context, thereby potentially limiting computerized LS analysis. In this work, LS and Impedance Pneumography (IP) measurements were obtained from 10 healthy volunteers while performing normal and forced-expiratory (FE) breathing maneuvers using our wearable IP and respiratory sounds (WIRS) system. Simultaneous auscultation was performed with the Eko CORE stethoscope (EKO). The breathing-phase context was extracted from the IP signals and used to compute phase-by-phase (Inspiratory (I), expiratory (E), and their ratio (I:E)) and breath-by-breath acoustic features. Their individual and added value was then elucidated through machine learning analysis. We found that the phase-contextualized features effectively captured the underlying acoustic differences between deep and FE breaths, yielding a maximum F1 Score of 84.1 ±11.4% with the phase-by-phase features as the strongest contributors to this performance. Further, the individual phase-contextualized models outperformed the traditional breath-by-breath models in all cases. The validity of the results was demonstrated for the LS obtained with WIRS, EKO, and their combination. These results suggest that incorporating breathing-phase context may enhance computerized LS analysis. Hence, multimodal sensing systems that enable this, such as WIRS, have the potential to advance LS clinical utility beyond traditional manual auscultation and improve patient care.

Exploring classical machine learning for identification of pathological lung auscultations

The use of machine learning in biomedical research has surged in recent years thanks to advances in devices and artificial intelligence. Our aim is to expand this body of knowledge by applying machine learning to pulmonary auscultation signals. Despite improvements in digital stethoscopes and attempts to find synergy between them and artificial intelligence, solutions for their use in clinical settings remain scarce. Physicians continue to infer initial diagnoses with less sophisticated means, resulting in low accuracy, leading to suboptimal patient care. To arrive at a correct preliminary diagnosis, the auscultation diagnostics need to be of high accuracy. Due to the large number of auscultations performed, data availability opens up opportunities for more effective sound analysis. In this study, digital 6-channel auscultations of 45 patients were used in various machine learning scenarios, with the aim of distinguishing between normal and abnormal pulmonary sounds. Audio features (such as fundamental frequencies F0-4, loudness, HNR, DFA, as well as descriptive statistics of log energy, RMS and MFCC) were extracted using the Python library Surfboard. Windowing, feature aggregation, and concatenation strategies were used to prepare data for machine learning algorithms in unsupervised (fair-cut forest, outlier forest) and supervised (random forest, regularized logistic regression) settings. The evaluation was carried out using 9-fold stratified cross-validation repeated 30 times. Decision fusion by averaging the outputs for a subject was also tested and found to be helpful. Supervised models showed a consistent advantage over unsupervised ones, with random forest achieving a mean AUC ROC of 0.691 (accuracy 71.11%, Kappa 0.416, F1-score 0.675) in side-based detection and a mean AUC ROC of 0.721 (accuracy 68.89%, Kappa 0.371, F1-score 0.650) in patient-based detection.

Detection of Mycoplasma pneumoniae in hospitalized children with pneumonia in Laos

Mycoplasma pneumoniae has been described worldwide as an important cause of community-acquired pneumonia. From December 2013 to December 2014, 461 children admitted to Mahosot Hospital, Vientiane, Laos, with acute respiratory infection were investigated for upper respiratory microorganisms using probe-based real-time polymerase chain reaction (PCR) (FTD33). M. pneumoniae was detected by FTD33 in the upper respiratory tract of three patients, two girls and one boy, 5.7 and 3.9 years old and 13.6 years old, respectively. They presented with clinical features compatible with M. pneumoniae infection. The two girls were also positive for other potential pathogens. The boy had abnormal pulmonary auscultation, and one of the girls had significant anaemia. These results suggest that enhancement of diagnostic systems for M. pneumoniae detection and analysis of its antibiotic resistance profile is needed to raise awareness and improve understanding of the epidemiology of M. pneumoniae infection in Laos, enable targeted therapy, and inform treatment guidelines.

Machine Learning for Automated Classification of Abnormal Lung Sounds Obtained from Public Databases: A Systematic Review.

Pulmonary auscultation is essential for detecting abnormal lung sounds during physical assessments, but its reliability depends on the operator. Machine learning (ML) models offer an alternative by automatically classifying lung sounds. ML models require substantial data, and public databases aim to address this limitation. This systematic review compares characteristics, diagnostic accuracy, concerns, and data sources of existing models in the literature. Papers published from five major databases between 1990 and 2022 were assessed. Quality assessment was accomplished with a modified QUADAS-2 tool. The review encompassed 62 studies utilizing ML models and public-access databases for lung sound classification. Artificial neural networks (ANN) and support vector machines (SVM) were frequently employed in the ML classifiers. The accuracy ranged from 49.43% to 100% for discriminating abnormal sound types and 69.40% to 99.62% for disease class classification. Seventeen public databases were identified, with the ICBHI 2017 database being the most used (66%). The majority of studies exhibited a high risk of bias and concerns related to patient selection and reference standards. Summarizing, ML models can effectively classify abnormal lung sounds using publicly available data sources. Nevertheless, inconsistent reporting and methodologies pose limitations to advancing the field, and therefore, public databases should adhere to standardized recording and labeling procedures.