Exploring the link between a novel approach for computer aided lung sound analysis and imaging biomarkers: a cross-sectional study.

BACKGROUND: Computer Aided Lung Sound Analysis (CALSA) aims to overcome limitations associated with standard lung auscultation by removing the subjective component and allowing quantification of sound characteristics. In this proof-of-concept study, a novel automated approach was evaluated in real patient data by comparing lung sound characteristics to structural and functional imaging biomarkers. METHODS: Patients with cystic fibrosis (CF) aged > 5y were recruited in a prospective cross-sectional study. CT scans were analyzed by the CF-CT scoring method and Functional Respiratory Imaging (FRI). A digital stethoscope was used to record lung sounds at six chest locations. Following sound characteristics were determined: expiration-to-inspiration (E/I) signal power ratios within different frequency ranges, number of crackles per respiratory phase and wheeze parameters. Linear mixed-effects models were computed to relate CALSA parameters to imaging biomarkers on a lobar level. RESULTS: 222 recordings from 25 CF patients were included. Significant associations were found between E/I ratios and structural abnormalities, of which the ratio between 200 and 400 Hz appeared to be most clinically relevant due to its relation with bronchiectasis, mucus plugging, bronchial wall thickening and air trapping on CT. The number of crackles was also associated with multiple structural abnormalities as well as regional airway resistance determined by FRI. Wheeze parameters were not considered in the statistical analysis, since wheezing was detected in only one recording. CONCLUSIONS: The present study is the first to investigate associations between auscultatory findings and imaging biomarkers, which are considered the gold standard to evaluate the respiratory system. Despite the exploratory nature of this study, the results showed various meaningful associations that highlight the potential value of automated CALSA as a novel non-invasive outcome measure in future research and clinical practice.

Breath Measurement Method for Synchronized Reproduction of Biological Tones in an Augmented Reality Auscultation Training System.

An educational augmented reality auscultation system (EARS) is proposed to enhance the reality of auscultation training using a simulated patient. The conventional EARS cannot accurately reproduce breath sounds according to the breathing of a simulated patient because the system instructs the breathing rhythm. In this study, we propose breath measurement methods that can be integrated into the chest piece of a stethoscope. We investigate methods using the thoracic variations and frequency characteristics of breath sounds. An accelerometer, a magnetic sensor, a gyro sensor, a pressure sensor, and a microphone were selected as the sensors. For measurement with the magnetic sensor, we proposed a method by detecting the breathing waveform in terms of changes in the magnetic field accompanying the surface deformation of the stethoscope based on thoracic variations using a magnet. During breath sound measurement, the frequency spectra of the breath sounds acquired by the built-in microphone were calculated. The breathing waveforms were obtained from the difference in characteristics between the breath sounds during exhalation and inhalation. The result showed the average value of the correlation coefficient with the reference value reached 0.45, indicating the effectiveness of this method as a breath measurement method. And the evaluations suggest more accurate breathing waveforms can be obtained by selecting the measurement method according to breathing method and measurement point.

Novel Approach Combining Shallow Learning and Ensemble Learning for the Automated Detection of Swallowing Sounds in a Clinical Database.

Cervical auscultation is a simple, noninvasive method for diagnosing dysphagia, although the reliability of the method largely depends on the subjectivity and experience of the evaluator. Recently developed methods for the automatic detection of swallowing sounds facilitate a rough automatic diagnosis of dysphagia, although a reliable method of detection specialized in the peculiar feature patterns of swallowing sounds in actual clinical conditions has not been established. We investigated a novel approach for automatically detecting swallowing sounds by a method wherein basic statistics and dynamic features were extracted based on acoustic features: Mel Frequency Cepstral Coefficients and Mel Frequency Magnitude Coefficients, and an ensemble learning model combining Support Vector Machine and Multi-Layer Perceptron were applied. The evaluation of the effectiveness of the proposed method, based on a swallowing-sounds database synchronized to a video fluorographic swallowing study compiled from 74 advanced-age patients with dysphagia, demonstrated an outstanding performance. It achieved an F1-micro average of approximately 0.92 and an accuracy of 95.20%. The method, proven effective in the current clinical recording database, suggests a significant advancement in the objectivity of cervical auscultation. However, validating its efficacy in other databases is crucial for confirming its broad applicability and potential impact.

SonicGuard Sensor-A Multichannel Acoustic Sensor for Long-Term Monitoring of Abdominal Sounds Examined through a Qualification Study.

Auscultation is a fundamental diagnostic technique that provides valuable diagnostic information about different parts of the body. With the increasing prevalence of digital stethoscopes and telehealth applications, there is a growing trend towards digitizing the capture of bodily sounds, thereby enabling subsequent analysis using machine learning algorithms. This study introduces the SonicGuard sensor, which is a multichannel acoustic sensor designed for long-term recordings of bodily sounds. We conducted a series of qualification tests, with a specific focus on bowel sounds ranging from controlled experimental environments to phantom measurements and real patient recordings. These tests demonstrate the effectiveness of the proposed sensor setup. The results show that the SonicGuard sensor is comparable to commercially available digital stethoscopes, which are considered the gold standard in the field. This development opens up possibilities for collecting and analyzing bodily sound datasets using machine learning techniques in the future.

Laennecs' mediate auscultation: Child's play.

Before Laennec, respiratory diseases that we recognize today were often confused, because the heart and lung are locked inside the rib cage. Impressed by the autopsies performed by Xavier Bichat (1771-1802), Laennec maintained the importance of the anatomoclinical method. But he indicated in his early 1820s lectures at the Collège de France that the discovery of auscultation was fortuitous and empirical. Duffin demonstrates that medical discoveries hardly obey an implacable logic, they arise outside of pre-established projects. In this paper, we retrace the chronology and antecedents at the origin of the important medical invention that is the stethoscope.

Reference blood pressure values obtained using the auscultation method for 2-year-old Japanese children: from the Japan Environment and Children's Study.

BACKGROUND: Reference blood pressure (BP) values for Japanese children based on a large number of measurements by auscultation have not yet been established. METHODS: This was a cross-sectional analysis of data from a birth-cohort study. The data from the sub-cohort study conducted for children at the age of 2 years in the Japan Environment and Children's Study from April 2015 to January 2017 were analyzed. BP was measured via auscultation using an aneroid sphygmomanometer. Each participant was measured in triplicate, and the average value of two consecutive measurements with a difference of less than 5 mmHg was recorded. The reference BP values were estimated using the lambda-mu-sigma (LMS) method and compared with those obtained via the polynomial regression model. RESULTS: Data from 3361 participants were analyzed. Although the difference between the estimated BP values by the LMS and the polynomial regression model was small, the LMS model was more valid based on the results of the fit curve of the observed values and regression models for each model. For 2-year-old children with heights in the 50th percentile, the 50th, 90th, 95th, and 99th percentile reference values of systolic BP (mmHg) for boys were 91, 102, 106, and 112, and that for girls were 90, 101, 103, and 109, respectively, and those of diastolic BP for boys were 52, 62, 65, and 71, and that for girls were 52, 62, 65, and 71, respectively. CONCLUSION: The reference BP values for 2-year-old Japanese children were determined based on auscultation and were made available.

Clinical tests for confirming tracheal intubation or excluding oesophageal intubation: a diagnostic test accuracy systematic review and meta-analysis.

Unrecognised oesophageal intubation causes preventable serious harm to patients undergoing tracheal intubation. When capnography is unavailable or doubted, clinicians still use clinical findings to confirm tracheal intubation, or exclude oesophageal intubation, and false reassurance from clinical examination is a recurring theme in fatal cases of unrecognised oesophageal intubation. We conducted a systematic review and meta-analysis of the diagnostic accuracy of five clinical examination tests and the oesophageal detector device when used to confirm tracheal intubation. We searched four databases for studies reporting index clinical tests against a reference standard, from inception to 28 February 2023. We included 49 studies involving 10,654 participants. Methodological quality was overall moderate to high. We looked at misting (three studies, 115 participants); lung auscultation (three studies, 217 participants); combined lung and epigastric auscultation (four studies, 506 participants); the oesophageal detector device (25 studies, 3024 participants); 'hang-up' (two non-human studies); and chest rise (one non-human study). The reference standards used were capnography (22 studies); direct vision (10 studies); and bronchoscopy (three studies). When used to confirm tracheal intubation, misting has a false positive rate (95%CI) of 0.69 (0.43-0.87); lung auscultation 0.14 (0.08-0.23); five-point auscultation 0.18 (0.08-0.36); and the oesophageal detector device 0.05 (0.02-0.09). Tests to exclude events that invariably lead to severe damage or death must have a negligible false positive rate. Misting or auscultation have too high a false positive rate to reliably exclude oesophageal intubation and there is insufficient evidence to support the use of 'hang-up' or chest rise. The oesophageal detector device may be considered where other more reliable means are not available, though waveform capnography remains the reference standard for confirmation of tracheal intubation.

Agreement of Oscillometric and Auscultatory blood pressure measurement methods: An ambulance noise simulation study.

OBJECTIVE: Although noise is known to negatively affect blood pressure (BP) measurements, its impact on different BP measurement methods remains unclear. The aim of this study is to compare the agreement of oscillometric and auscultatory BP measurement methods under in-ambulance noise levels. METHODS: This method-comparison study was conducted on 50 healthy volunteers in a tertiary emergency department (ED). Participants were divided into two groups of 25, and BP was measured using auscultatory and oscillometric methods in noisy and ambient environments by 2 emergency medicine technicians (EMT). The primary object of the study was to compare the agreement of auscultatory mercury sphygmomanometers and automated auscillometric BP measurements in ambient and noisy environments. RESULTS: We examined the agreement between auscultative and oscillometric measurements of BP conducted in an ambient environment (46.75 [IQR (41.2--55.18)] dB) and found that both systolic and diastolic BP were within the level of agreement (LoA) established before the study (systolic BP [-13.96 to 8.48 mmHG], diastolic BP [-7.44 to 8.08 mmHg]); whereas, in noisy environment (92.35 [IQR 88-96.55] dB) both systolic and diastolic BP were outside the range of LoA (systolic BP [-37.77 to 9.94 mmHg], diastolic BP [-21.73 to 16.37 mmHg]). Additionally, we found that in ambient environments, concordance correlation coefficients were higher than in noisy environments (0.943 [0.906-0.966], 0.957 [0.93-0.974]; 0.574 [0.419-0.697], 0.544 [0.326-0.707]; systolic and diastolic BP, respectively). CONCLUSION: The results of this study demonstrate that noise significantly affects the agreement between oscillometric and auscultatory blood pressure measurement methods.

Clinical and imaging factors that can predict contagiousness of pulmonary tuberculosis.

BACKGROUND: Knowledge on predicting pulmonary tuberculosis (PTB) contagiosity in the hospital admission setting is limited. The objective was to assess clinical and radiological criteria to predict PTB contagiosity. METHODS: Retrospective analysis of 7 clinical, 4 chest X-ray (CXR) and 5 computed tomography (CT) signs in 299 PTB patients admitted to an urban tertiary hospital from 2008 to 2016. If the acid fact bacilli stain was positive (AFB+) on admission, the case was considered high contagiosity. RESULTS: Best predictors for high PTB contagiosity (AFB+) were haemoptysis (OR 4.33), cough (3.00), weight loss (2.96), cavitation in CT (2.75), cavitation in CXR (2.55), tree-in-bud-sign in CT (2.12), German residency of the patient (1.89), and abnormal auscultation findings (1.83). A previous TB infection reduced the risk of contagiosity statistically (0.40). Radiographic infiltrates, miliary picture, and pleural effusion were not helpful in predicting high or low contagiosity. 34% of all patients were clinically asymptomatic (20% of the highly contagious group, 50% of the low contagious group). CONCLUSION: Haemoptysis, cough and weight loss as well as cavitation and tree-in-bud sign in CXR/CT can be helpful to predict PTB contagiosity and to improve PTB management.

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.

Performance of the Aktiia optical blood pressure measurement device in the elderly: a comparison with double blinded auscultation in different body positions.

BACKGROUND: Accurate blood pressure (BP) measurement is essential for the correct diagnosis and management of hypertension (HTN) especially in the elderly population. As with of all BP devices, the accuracy of cuffless devices must be verified. This study (NCT04027777) aimed to evaluate the performance of a wrist cuffless optical BP device in an elderly population cohort in different body positions with auscultation as the reference measurement. DESIGN AND METHODS: Patients aged 65-85 years with different BP categories but without diabetes were recruited. After an initial calibration based on auscultatory measurements, BP estimation from the Aktiia Bracelet (Aktiia SA, Switzerland) were compared to reference double-blinded auscultatory measurements in sitting, standing and lying positions on four separate visits distributed over one month. In the absence of a universal standard for cuffless BP device at the time of the study, modified ISO81060-2 criteria were used for performance analysis. RESULTS: Thirty-five participants were included in the analysis fulfilling the inclusion requirements of ISO 81060-2. A total of 469 paired measurements were obtained with overall 83% acceptance rate. Differences (mean ± SD)   between Aktiia Bracelet and auscultation for systolic BP were -0.26 ± 9.96 mmHg for all body positions aggregated (sitting 1.23 ± 7.88 mmHg, standing -1.81 ± 11.11 mmHg, lying -1.8 ± 9.96 mmHg). Similarly, differences for diastolic BP were -0.75 ± 7.0 mmHg (0.2 ± 5.55 mmHg, -5.35 ± 7.75 mmHg and -0.94 ± 7.47 mmHg, respectively). Standard deviation of the averaged differences per subject for systolic/diastolic BP was 3.8/2.5 mmHg in sitting and 4.4/3.7 mmHg for all body positions aggregated. CONCLUSIONS: Overall, this study demonstrates a similar performance of the Aktiia Bracelet compared to auscultation in an elderly population in body positions representative of daily activities. The use of more comfortable, non-invasive, and non-occlusive BP monitors during long periods may facilitate e-health and may contribute to better management of HTN, including diagnosis and treatment of HTN, in the elderly.

Accuracy of blood pressure measurements is essential in the diagnosis and the follow-up of patients with high blood pressure. As with any blood pressure measuring device, a validation is necessary. In this study including a elderly population, we compared values obtained by the cuffless Aktiia Bracelet (Aktiia SA, Switzerland) after an initial calibration with the reference auscultatory method during four separate study days distributed over one month. We show that the accuracy of the Aktiia Bracelet is similar to auscultation. The accuracy varies depending on the position in which the measurement is performed. Overall, the accuracy is not modified by a higher age category. The use of a cuffless device in the elderly population characterized by high prevalence of hypertension may facilitate the follow-up of blood pressure with more comfort and minimal constraints.

Dysphagia Management and Cervical Auscultation: Reliability and Validity Against FEES.

This study investigated the reliability and validity (sensitivity and specificity) of cervical auscultation (CA) using both swallow and pre-post swallow-respiratory sounds, as compared with Flexible Endoscopic Evaluation of Swallowing (FEES). With 103 swallow-respiratory sequences from 23 heterogenic patients, these swallows sounds were rated by eight CA-trained Speech-Language Pathologists (SLPs) to investigate: (1) if the swallow was safe (primary outcome); (2) patient dysphagia status; (3) the influence of liquid viscosity on CA accuracy (secondary outcomes). Primary outcome data showed high CA sensitivity (85.4%), and specificity (80.3%) with all consistencies for the safe measurement, with CA predictive values of [Formula: see text] 90% to accurately detect unsafe swallows. Intra-rater reliability was good (Kappa [Formula: see text] 0.65), inter rater reliability moderate (Kappa [Formula: see text] 0.58). Secondary outcome measures showed high sensitivity (80.1%) to identify if a patient was dysphagic, low specificity (22.9%), and moderate correlation (rs [Formula: see text] 0.62) with FEES. A difference across bolus viscosities identified that CA sensitivities (90.1%) and specificities ([Formula: see text] 84.7%) for thin liquids were greater than for thick liquids (71.0-77.4% sensitivities, 74.0-81.3% specificities). Results demonstrate high validity and moderate-good reliability of CA-trained SLPs to determine swallow safety when compared with FEES. Data support the use of CA as an adjunct to the clinical swallow examination. CA should include pre-post respiratory sounds and requires specific training. Clinical implications: The authors advocate for holistic dysphagia management including instrumental assessment and ongoing CSE/review [Formula: see text] CA. Adding CA to the CSE/review does not replace instrumental assessment, nor should CA be used as a stand-alone tool.

Masking of Korotkoff sounds used in blood pressure measurement through auscultation.

Measurement of blood pressure (BP) through manual auscultation and the observation of Korotkoff sounds (KSs) remains the gold standard in BP methodology. Critical to determining BP levels via auscultation is the determination of KS audibility. While absolute sound level audibility is well researched, the problem has not been approached from the point of view of psychoacoustic masking of the sounds. Here, during manual auscultation of BP, a direct comparison is made between what an observer perceives as audible and the electronic analysis of audibility level determined from masking of sound signal levels. KSs are collected during auscultation with an electronic stethoscope, which allows simultaneously observing sound audibility and recording the sound electronically. By time-segmenting the recorded sound around Korotkoff peaks into a test segment and a masking segment, performing Fourier transforms on the segments, and comparing frequency-band sound energy levels, signal-to-noise ratios of a sound to its masking counterpart can be defined. Comparing these ratios to difference limen in the psychoacoustic masking literature, an approximate threshold for sound audibility is obtained. It is anticipated that this approach could have profound effects on future development of automated auscultation BP measurements.

Cardiac screening in school children: Combining auscultation and electrocardiography with a crowdsourcing model.

BACKGROUND/PURPOSE: School-based cardiac screening is useful for identifying children and adolescents with a high risk of sudden cardiac death. However, because of challenges associated with cost, distance, and human resources, cardiac screening is not widely implemented, especially in rural areas with limited medical resources. This study aims to establish a cloud-based system suitable for mass cardiac screening of schoolchildren in rural areas with limited medical resources. METHODS: Students from three schools were included. They or their guardians completed a simple questionnaire, administered in paper or electronic form. Heart sounds were recorded using an electronic stethoscope. Twelve-lead electrocardiograms (ECGs) were recorded and digitalized. The signals were transmitted through Bluetooth to a tablet computer and then uploaded to a cloud server over Wi-Fi. Crowdsourced pediatric cardiologists reviewed those data from a web-based platform and provided remote consultation. In cases in which abnormal heart sounds or ECGs were noted, the students were referred to the hospital for further evaluation. RESULTS: A total of 1004 students were enrolled in this study. Of the 138 students referred, 62 were diagnosed as having an abnormal heart condition and most had previously been undiagnosed. The interrater agreeability was high. CONCLUSION: An innovative strategy combining a cloud-based cardiac screening system with remote consultation by crowdsourced experts was established. This system allows pediatric cardiologists to provide consultation and make reliable diagnoses. Combined with crowdsourcing, the system constitutes a viable approach for mass cardiac screening in children and adolescents living in rural areas with insufficient medical resources.

Constant Force-Tracking Control Based on Deep Reinforcement Learning in Dynamic Auscultation Environment.

Intelligent medical robots can effectively help doctors carry out a series of medical diagnoses and auxiliary treatments and alleviate the current shortage of social personnel. Therefore, this paper investigates how to use deep reinforcement learning to solve dynamic medical auscultation tasks. We propose a constant force-tracking control method for dynamic environments and a modeling method that satisfies physical characteristics to simulate the dynamic breathing process and design an optimal reward function for the task of achieving efficient learning of the control strategy. We have carried out a large number of simulation experiments, and the error between the tracking of normal force and expected force is basically within ±0.5 N. The control strategy is tested in a real environment. The preliminary results show that the control strategy performs well in the constant force-tracking of medical auscultation tasks. The contact force is always within a safe and stable range, and the average contact force is about 5.2 N.

Heart Sound Classification Network Based on Convolution and Transformer.

Electronic auscultation is vital for doctors to detect symptoms and signs of cardiovascular diseases (CVDs), significantly impacting human health. Although progress has been made in heart sound classification, most existing methods require precise segmentation and feature extraction of heart sound signals before classification. To address this, we introduce an innovative approach for heart sound classification. Our method, named Convolution and Transformer Encoder Neural Network (CTENN), simplifies preprocessing, automatically extracting features using a combination of a one-dimensional convolution (1D-Conv) module and a Transformer encoder. Experimental results showcase the superiority of our proposed method in both binary and multi-class tasks, achieving remarkable accuracies of 96.4%, 99.7%, and 95.7% across three distinct datasets compared with that of similar approaches. This advancement holds promise for enhancing CVD diagnosis and treatment.

Suppression of Clothing-Induced Acoustic Attenuation in Robotic Auscultation.

For patients who are often embarrassed and uncomfortable when exposing their breasts and having them touched by physicians of different genders during auscultation, we are developing a robotic system that performs auscultation over clothing. As the technical issue, the sound obtained through the clothing is often attenuated. This study aims to investigate clothing-induced acoustic attenuation and develop a suppression method for it. Because the attenuation is due to the loss of energy as sound propagates through a medium with viscosity, we hypothesized that the attenuation is improved by compressing clothing and shortening the sound propagation distance. Then, the amplitude spectrum of the heart sound was obtained over clothes of different thicknesses and materials in a phantom study and human trial at varying contact forces with a developed passive-actuated end-effector. Our results demonstrate the feasibility of the attenuation suppression method by applying an optimum contact force, which varied according to the clothing condition. In the phantom experiments, the attenuation rate was improved maximumly by 48% when applying the optimal contact force (1 N). In human trials, the attenuation rate was under the acceptable attenuation (40%) when applying the optimal contact force in all combinations in each subject. The proposed method promises the potential of robotic auscultation toward eliminating gender bias.

Improving Valvular Pathologies and Ventricular Dysfunction Diagnostic Efficiency Using Combined Auscultation and Electrocardiography Data: A Multimodal AI Approach.

Simple sensor-based procedures, including auscultation and electrocardiography (ECG), can facilitate early diagnosis of valvular diseases, resulting in timely treatment. This study assessed the impact of combining these sensor-based procedures with machine learning on diagnosing valvular abnormalities and ventricular dysfunction. Data from auscultation at three distinct locations and 12-lead ECGs were collected from 1052 patients undergoing echocardiography. An independent cohort of 103 patients was used for clinical validation. These patients were screened for severe aortic stenosis (AS), severe mitral regurgitation (MR), and left ventricular dysfunction (LVD) with ejection fractions ≤ 40%. Optimal neural networks were identified by a fourfold cross-validation training process using heart sounds and various ECG leads, and their outputs were combined using a stacking technique. This composite sensor model had high diagnostic efficiency (area under the receiver operating characteristic curve (AUC) values: AS, 0.93; MR, 0.80; LVD, 0.75). Notably, the contribution of individual sensors to disease detection was found to be disease-specific, underscoring the synergistic potential of the sensor fusion approach. Thus, machine learning models that integrate auscultation and ECG can efficiently detect conditions typically diagnosed via imaging. Moreover, this study highlights the potential of multimodal artificial intelligence applications.

StethAid: A Digital Auscultation Platform for Pediatrics.

(1) Background: Mastery of auscultation can be challenging for many healthcare providers. Artificial intelligence (AI)-powered digital support is emerging as an aid to assist with the interpretation of auscultated sounds. A few AI-augmented digital stethoscopes exist but none are dedicated to pediatrics. Our goal was to develop a digital auscultation platform for pediatric medicine. (2) Methods: We developed StethAid-a digital platform for artificial intelligence-assisted auscultation and telehealth in pediatrics-that consists of a wireless digital stethoscope, mobile applications, customized patient-provider portals, and deep learning algorithms. To validate the StethAid platform, we characterized our stethoscope and used the platform in two clinical applications: (1) Still's murmur identification and (2) wheeze detection. The platform has been deployed in four children's medical centers to build the first and largest pediatric cardiopulmonary datasets, to our knowledge. We have trained and tested deep-learning models using these datasets. (3) Results: The frequency response of the StethAid stethoscope was comparable to those of the commercially available Eko Core, Thinklabs One, and Littman 3200 stethoscopes. The labels provided by our expert physician offline were in concordance with the labels of providers at the bedside using their acoustic stethoscopes for 79.3% of lungs cases and 98.3% of heart cases. Our deep learning algorithms achieved high sensitivity and specificity for both Still's murmur identification (sensitivity of 91.9% and specificity of 92.6%) and wheeze detection (sensitivity of 83.7% and specificity of 84.4%). (4) Conclusions: Our team has created a technically and clinically validated pediatric digital AI-enabled auscultation platform. Use of our platform could improve efficacy and efficiency of clinical care for pediatric patients, reduce parental anxiety, and result in cost savings.

A Low-Cost AI-Empowered Stethoscope and a Lightweight Model for Detecting Cardiac and Respiratory Diseases from Lung and Heart Auscultation Sounds.

Cardiac and respiratory diseases are the primary causes of health problems. If we can automate anomalous heart and lung sound diagnosis, we can improve the early detection of disease and enable the screening of a wider population than possible with manual screening. We propose a lightweight yet powerful model for simultaneous lung and heart sound diagnosis, which is deployable in an embedded low-cost device and is valuable in remote areas or developing countries where Internet access may not be available. We trained and tested the proposed model with the ICBHI and the Yaseen datasets. The experimental results showed that our 11-class prediction model could achieve 99.94% accuracy, 99.84% precision, 99.89% specificity, 99.66% sensitivity, and 99.72% F1 score. We designed a digital stethoscope (around USD 5) and connected it to a low-cost, single-board-computer Raspberry Pi Zero 2W (around USD 20), on which our pretrained model can be smoothly run. This AI-empowered digital stethoscope is beneficial for anyone in the medical field, as it can automatically provide diagnostic results and produce digital audio records for further analysis.