Auscultation Points Research Articles

Murmur identification and outcome prediction in phonocardiograms using deep features based on Stockwell transform

Traditionally, heart murmurs are diagnosed through cardiac auscultation, which requires specialized training and experience. The purpose of this study is to predict patients' clinical outcomes (normal or abnormal) and identify the presence or absence of heart murmurs using phonocardiograms (PCGs) obtained at different auscultation points. A semi-supervised model tailored to PCG classification is introduced in this study, with the goal of improving performance using time–frequency deep features. The study begins by investigating the behavior of PCGs in the time–frequency domain, utilizing the Stockwell transform to convert the PCG signal into two-dimensional time–frequency maps (TFMs). A deep network named AlexNet is then used to derive deep feature sets from these TFMs. In feature reduction, redundancy is eliminated and the number of deep features is reduced to streamline the feature set. The effectiveness of the extracted features is evaluated using three different classifiers using the CinC/Physionet challenge 2022 dataset. For Task I, which focuses on heart murmur detection, the proposed approach achieved an average accuracy of 93%, sensitivity of 91%, and F1-score of 91%. According to Task II of the CinC/Physionet challenge 2022, the approach showed a clinical outcome cost of 5290, exceeding the benchmark set by leading methods in the challenge.

AUGMENTED REALITY APPLICATION FOR PULMONARY AUSCULTATION LEARNING AID

AbstractDuring some applied anatomy lessons, physiotherapist students must develop a skill of locating and recognizing underlying anatomical elements, from surface palpations, or through clinical tools on their classmates. For this, precise procedures exist and students have to mobilize knowledge acquired from different types of resources: support documents from the teacher, anatomical charts (books, drawings, diagrams), dissection videos or even internet resources. They also need more practical time and exercises. The issues facing students relate to the availability of these resources during the procedure, their relevance, the mobilisation of resources to practice and the applicability in a real situation.This project stems from a desire to make Augmented Reality technology available to students, to help in learning precise clinical procedure, in our case the location of the auscultation points using a stethoscope. The aim would be to supplement the real environment of the student during his learning on a person, by the possibilities of having additional resources according to its needs (auscultation points superimposed on a 3D representation of the lungs).

Low‐Intensity Sensitive and High Stability Flexible Heart Sound Sensor Enabled by Hybrid Near‐Field/Far‐Field Electrospinning

AbstractLow‐cost and wearable heart sound sensors can facilitate early detection and monitoring of cardiovascular and respiratory diseases. Such sensors are currently limited by either the complexity of fabrication processes or low sensitivity and reliability for weak signal detection. Here, a hybrid near‐field/far‐field electrospinning approach is demonstrated that enables low‐cost fabrication and optimization of triboelectric heterostructures for heart sound sensing. Specifically, by combining the far‐field produced highly polarized and porous polyvinylidene difluoride network for triboelectric electrification and near‐field patterned polyurethane grid spacers for vibration enhancement and charge trapping, the greatly improved sensor output at the heart sound frequency (50–150 Hz) and intensity (<80 dB) range, demonstrating record high sensitivity of 7027 mV Pa−1 and low detection limit of 47 dB. The sensor exhibits excellent stability under both sudden physical disturbance and long‐term cycling stress, showing no degradation during 7 h of continuous operation. It is demonstrated that the sensor can be integrated with apparel and capture high‐quality heart sound signals at all five diagnostic auscultation points and distinguish characteristic heart sound patterns caused by different cardiovascular diseases.

Development of a Collaborative Robotic Platform for Autonomous Auscultation

Listening to internal body sounds, or auscultation, is one of the most popular diagnostic techniques in medicine. In addition to being simple, non-invasive, and low-cost, the information it offers, in real time, is essential for clinical decision-making. This process, usually done by a doctor in the presence of the patient, currently presents three challenges: procedure duration, participants’ safety, and the patient’s privacy. In this article we tackle these by proposing a new autonomous robotic auscultation system. With the patient prepared for the examination, a 3D computer vision sub-system is able to identify the auscultation points and translate them into spatial coordinates. The robotic arm is then responsible for taking the stethoscope surface into contact with the patient’s skin surface at the various auscultation points. The proposed solution was evaluated to perform a simulated pulmonary auscultation in six patients (with distinct height, weight, and skin color). The obtained results showed that the vision subsystem was able to correctly identify 100% of the auscultation points, with uncontrolled lighting conditions, and the positioning subsystem was able to accurately position the gripper on the corresponding positions on the human body. Patients reported no discomfort during auscultation using the described automated procedure.

Quantifying the Effect of Body Habitus on Cardiac Auscultation Via Computational Hemoacoustics.

The effect of body habitus on auscultation of heart murmurs is investigated via computational hemoacoustic modeling. The source of the heart murmur is first obtained from a hemodynamic simulation of blood flow through a stenosed aortic valve. This sound source is then placed at the aortic valve location in four distinct human thorax models, and the propagation of the murmur in each thorax model is simulated by solving the elastic wave equations in the time-domain. Placing the same sound source in different thorax models allows for the disambiguation of the effect of body habitus on cardiac auscultation. The surface acceleration resulting from the murmur on each subject's chest surface shows that subjects with higher body-mass index and thoracic cross-sectional area yield smaller acceleration values for the S1 sound. Moreover, the spectral analysis of the signal shows that slope from linear regression in the normal heart sound frequency range (10-150 Hz) is larger for children at the aortic, pulmonic, and mitral auscultation points compared to that for adults. The slope in the murmur frequency range (150-400 Hz) was larger for female subjects at the mitral point compared to that for male subjects. The trends from the results show the potential of the proposed computational method to provide quantitative insights regarding the effect of various anatomical factors on cardiac auscultation.

Artificial Intelligence-Based Stethoscope for the Diagnosis of Aortic Stenosis

BackgroundThe diagnostic accuracy of the stethoscope is limited and highly dependent on clinical expertise. Our purpose was to develop an electronic stethoscope, based on artificial intelligence (AI) and infrasound, for the diagnosis of aortic stenosis (AS). MethodsWe used an electronic stethoscope (VoqX; Sanolla, Nesher, Israel) with subsonic capabilities and acoustic range of 3-2000 Hz. The study had 2 stages. In the first stage, using the VoqX, we recorded heart sounds from 100 patients referred for echocardiography (derivation group), 50 with moderate or severe AS and 50 without valvular disease. An AI-based supervised learning model was applied to the auscultation data from the first 100 patients used for training, to construct a diagnostic algorithm that was then tested on a validation group (50 other patients, 25 with AS and 25 without AS). In the second stage, conducted at a different medical center, we tested the device on 106 additional patients referred for echocardiography, which included patients with other valvular diseases. ResultsUsing data collected at the aortic and pulmonic auscultation points from the derivation group, the AI-based algorithm identified moderate or severe AS with 86% sensitivity and 100% specificity. When applied to the validation group, the sensitivity was 84% and specificity 92%; and in the additional testing group, 90% and 84%, respectively. The sensitivity was 55% for mild, 76% for moderate, and 93% for severe AS. ConclusionOur initial findings show that an AI-based stethoscope with infrasound capabilities can accurately diagnose AS. AI-based electronic auscultation is a promising new tool for automatic screening and diagnosis of valvular heart disease.

Artificial intelligence-based stethoscope for the diagnosis of aortic stenosis

Abstract Background Although the stethoscope is used for auscultation in the diagnosis of valvular heart disease for more than 200 years, its diagnostic accuracy is limited and highly dependent on clinical expertise and acoustic range of the human ear. Purpose To develop an electronic stethoscope, based on artificial intelligence (AI), for the diagnosis of aortic stenosis (AS). Methods We developed an electronic stethoscope (VoqxTM, Sanolla) with subsonic capabilities and acoustic range of 0–2,000 Hz. Using the VoqxTM, we recorded heart sounds from 100 patients referred for echocardiography (derivation group), 50 with moderate or severe AS (aortic valve area (AVA) ≤1.5 cm2) and 50 without valvular disease, using the 5 standard auscultation points. An AI based supervised learning model was applied to the auscultation data from the first 100 patients, to construct a diagnostic algorithm that was then tested on a validation group (50 other patients, 25 with AS and 25 without AS). Results The derivation group included 50 patients with AS (age 78±9 years, 28 males, AVA=0.87±0.19 cm2, mean gradient 44±16 mmHg, 39 with severe AS) and 50 without AS (age 58±16 years, 39 males). The AI based algorithm using 1–4 auscultation position (all except the mitral position), correctly identified 47/50 AS patients (sensitivity 94%) and 49/50 patients without AS (specificity 98%), total accuracy 96% (Table). The algorithm was then applied to the validation group (AS: n=25, age 76±9 years,12 males, AVA=0.88±0.21 cm2, mean gradient 41±15 mmHg, 18 with severe AS; No AS: n=25, age = 59±18 years, 17 males). The algorithm correctly identified 21/25 AS patients (sensitivity 84%) and 24/25 patients without AS (specificity 96%), total accuracy 90% (Table 1). Conclusion Our initial findings show that an AI based stethoscope can accurately diagnose AS. AI based electronic auscultation is a promising new tool for automatic diagnosis of valvular heart disease. Funding Acknowledgement Type of funding sources: Private company. Main funding source(s): Sanolla Company Table 1

A Framework for AI-Assisted Detection of Patent Ductus Arteriosus from Neonatal Phonocardiogram

The current diagnosis of Congenital Heart Disease (CHD) in neonates relies on echocardiography. Its limited availability requires alternative screening procedures to prioritise newborns awaiting ultrasound. The routine screening for CHD is performed using a multidimensional clinical examination including (but not limited to) auscultation and pulse oximetry. While auscultation might be subjective with some heart abnormalities not always audible it increases the ability to detect heart defects. This work aims at developing an objective clinical decision support tool based on machine learning (ML) to facilitate differentiation of sounds with signatures of Patent Ductus Arteriosus (PDA)/CHDs, in clinical settings. The heart sounds are pre-processed and segmented, followed by feature extraction. The features are fed into a boosted decision tree classifier to estimate the probability of PDA or CHDs. Several mechanisms to combine information from different auscultation points, as well as consecutive sound cycles, are presented. The system is evaluated on a large clinical dataset of heart sounds from 265 term and late-preterm newborns recorded within the first six days of life. The developed system reaches an area under the curve (AUC) of 78% at detecting CHD and 77% at detecting PDA. The obtained results for PDA detection compare favourably with the level of accuracy achieved by an experienced neonatologist when assessed on the same cohort.

About Possibility of Remote Diagnostics of the Respiratory System by Auscultation

Development of technical base, software, accumulated information on the diagnosis of the respiratory system provided the prerequisites for creating remote diagnostics of the human respiratory system through auscultation. The known methods do not solve the problem of determining auscultation points at patent´s housing without a diagnostic specialist. The purpose of this study is to develop a method for remote diagnostics of the respiratory system which provides ability to determine the points of auscultation without presence of a diagnostic specialist.The definition of auscultation points is provided using a computer program that allows to calculate the points´ coordinates based on the coordinates of points that determine the anatomical structure of the patient's torso. The patient or his assistant places the recording device at the auscultation points combining their images on the display with the image of the location of the recording device. The signal recorded at the auscultation point is remotely transmitted to a specialist for direct analysis and/or computer processing. The diagnostic module consists of two main units. The first unit contains a stethoscope, microphone, and amplifier connected to a mobile phone or other similar device containing an accelerometer. The patient or his assistant at the housing uses the unit. The second unit is a mobile phone with a mechanical marker or a computer with the ability to access the network in conjunction with the necessary software and is used remotely by a diagnostic specialist. The layout of the unit for recording and transmitting breath sounds was made. To avoid discrepancies in the diagnostic results the technical characteristics of the module elements must be normalized. Unified software is required for the module to function. The organizational tasks that need to be solved for the implementation of diagnostics are formulated.Use of the method of remote diagnostics of the respiratory system, providing the ability to determine points of auscultation without the direct presence of a diagnostic specialist and the module will allow increasing efficiency of treatment of pulmonary diseases reduce infection risks and economic costs.

A Robust and Real-Time Capable Envelope-Based Algorithm for Heart Sound Classification: Validation under Different Physiological Conditions.

This paper proposes a robust and real-time capable algorithm for classification of the first and second heart sounds. The classification algorithm is based on the evaluation of the envelope curve of the phonocardiogram. For the evaluation, in contrast to other studies, measurements on 12 probands were conducted in different physiological conditions. Moreover, for each measurement the auscultation point, posture and physical stress were varied. The proposed envelope-based algorithm is tested with two different methods for envelope curve extraction: the Hilbert transform and the short-time Fourier transform. The performance of the classification of the first heart sounds is evaluated by using a reference electrocardiogram. Overall, by using the Hilbert transform, the algorithm has a better performance regarding the F1-score and computational effort. The proposed algorithm achieves for the S1 classification an F1-score up to 95.7% and in average 90.5%. The algorithm is robust against the age, BMI, posture, heart rate and auscultation point (except measurements on the back) of the subjects.

Design and Evaluation of a Diaphragm for Electrocardiography in Electronic Stethoscopes.

Combining Phonocardiography (PCG) and Electrocardiography (ECG) data has been recognized within the state-of-the-art as of added value for enhanced cardiovascular assessment. However, multiple aspects of ECG data acquisition in a stethoscope form factor remain unstudied, and existing devices typically enforce a substantial change into routine clinical auscultation procedures, with predictably low technology acceptance. As such, in this paper, we present a novel approach to ECG data acquisition throughout the five main cardiac auscultation points, and that intends to be incorporated in a commonly used electronic stethoscope. Therefore, it enables analysis and acquisition of both PCG and ECG signals in a single pass. We describe the development, experimental evaluation, and comparison of the ECG signals obtained using our proposed approach and a gold standard medical device, through metrics that allow the evaluation of morphological similarities. Results point to a high correlation between the two evaluated setups, thus supporting the idea of meaningfully collecting ECG data along medical auscultation points with the proposed form factor. Moreover, this work has led us to conclude that for the studied population, signals acquired on focuses F1, F2, and F3 are usually highly correlated with leads V1 and V2 of the standard ECG medical recording procedure.

Cardiological «patterns» in term newborn with no symptoms

Clinical examination is essential for detection of cardiac pathology in term newborns. Sensitivity of cardiac noise estimation for diagnostics of congenital malformations is up to 88%. Detection of murmurs is quite subjective and depends on the skill and experience of the doctor. Usage of an electronic stethoscope with combination of digital recording of cardiac sounds and computer analysis significantly expands the possibilities of auscultation in the diagnostics of structural heart anomalies diagnostics.Purpose.Study of the structure of echocardiographic symptoms in clinically healthy newborns in comparison with the electronic auscultation analysis data.Materials and methods. A total of 195 healthy term newborns were examined, in which no structural features in the heart and large vessels were prenatally found. No pathological changes were identified with the traditional auscultation after birth. During the first 5 days of life, doppler echocardiography, differential pulse-oximetry, electronic auscultation were performed. The analysis of the received phonocardiograms was carried out with the developed computer program "Hearttone-D" with the analysis of 17 parameters of tone I and II and 16 parameters of intervals between tones at five standard listening points.Results. During the Doppler echocardiography study, all children had a functional oval window measuring from 2.0 to 4.5 mm. In 5 children (2.6%) the small size defects in the interventricular membrane were found. Signs of the PDA during the observation period were found in 54 (27.6%) newborns. Pulse-oximetry results in newborns were within 96-99% range.For the computer analysis, 27 phonocardiographs of newborns with PDA and 28 neonatal phonocadiograms of newborns with closed ductus arteriosus were selected. The inclusion criterion was the size of a functioning oval window 2.5-3.0 mm, the absence of any structural anomalies of the heart, the quality of recording of the phonocardiograph (no exterior sounds). The largest statistically significant difference between the parameters was observed at the II auscultation point (the maximum amplitude of tone I, the average value of tone II, the average amplitude module in the first interval, the mean amplitude module in the last quarter of the second interval, the width of the second interval).Conclusions. Тhe presence of clinically undetectable at the time of the PDA is detected by electronic auscultation end the highest statistical significance is characterized by parameters registered in the II point of auscultation.

Auf den Spuren des Erb’schen Auskultationspunkts: Rätsel gelöst

"Erb's point" is the fifth point of auscultation for the heart exam, located in the third intercostal space close to the sternum. It has sometimes been attributed to famous German neurologist Wilhelm Heinrich Erb (1840 - 1921), but without historical evidence. Erb's focus on neurology suggested that the auscultation point may have been confused with other points in the neck named after Erb. As Erb was a specialist for neurological manifestations of syphilis, we speculated that the heart murmur of aortic incompetence produced by syphilitic aortitis, best heard at Erb's point, linked Erb's name to the auscultation point. However, we eventually found a publication by an American physician who visited Erb's lectures in Heidelberg and reported that Erb explicitly introduced this "fifth point" in his case presentations in the late 1890 s. After all, Erb was chair of general medicine, with the first German chair of neurology only being established in 1919.

HUMAN IDENTIFICATION BASED ON HEART SOUND AUSCULTATION POINT

The application of human identification and verification has widely been used for over the past few decades. Drawbacks of such system however, are inevitable as forgery sophisticatedly developed alongside the technology advancement. Thus, this study proposed a research on the possibility of using heart sound as biometric. The main aim is to find an optimal auscultation point of heart sounds from either aortic, pulmonic, tricuspid or mitral that will most suitable to be used as the sound pattern for personal identification. In this study, the heart sound was recorded from 92 participants using a Welch Allyn Meditron electronic stethoscope whereas Meditron Analyzer software was used to capture the signal of heart sounds and ECG simultaneously for duration of 1 minute. The system is developed by a combination Mel Frequency Cepstrum Coefficients (MFCC) and Hidden Markov Model (HMM). The highest recognition rate is obtained at aortic area with 98.7% when HMM has 1 state and 32 mixtures, the lowest Equal Error Rate (EER) achieved was 0.9% which is also at aortic area. In contrast, the best average performance of HMM for every location is obtained at mitral area with 99.1% accuracy and 17.7% accuracy of EER at tricuspid area.

Detection of patients considering observation frequency of continuous and discontinuous adventitious sounds in lung sounds.

We propose an improved approach for distinguishing between healthy subjects and patients with pulmonary emphysema by the use of one stochastic acoustic model for continuous adventitious sounds and another for discontinuous adventitious sounds. These models are able to represent the spectral features of the adventitious sounds for the detection of abnormal respiration. However, abnormal respiratory sounds with unclassifiable spectral features are present among the continuous and discontinuous adventitious sounds and mixing noises. These sounds cause difficulties in achieving a highly accurate classification. In this study, the difference in occurrence frequencies between two types of adventitious sounds for each considered auscultation point and inspiration/expiration was considered. This difference, in combination with the confusion tendency of the classifier, was formulated as the validity score of each respiratory sound. The conventional spectral likelihood and the newly formulated validity score were combined to perform detection of abnormal respiration and patients. In the classification of healthy subjects and patients, the proposed approach achieved a higher classification rate (87.7%) than the conventional method (85.2%), demonstrating the statistical superiority (5% level) of the former.

Cardiac auscultation: Preliminary findings of a pilot study using continuous Wave Doppler and comparison with classic auscultation

Routine fetal heart examination is performed in the obstetric office using a portable continuous Wave Doppler (CWD) device [ [1] Brown A.D. Robertson J.G. The ultrasonic Doppler cardioscope in obstetrics. J Obstet Gynaecol Br Commonw. 1968 Jan; 75: 92-96 Crossref PubMed Scopus (5) Google Scholar ], which produces sounds but not images. We explored the heart in adult patients using this method and we found it to be surprisingly informative. We decided then to define cardiac auscultation points and typical auditive patterns for the procedure. After that, we compared this method with classic auscultation (CA) in cardiologic patients, using color Doppler echocardiography (CDE) as a gold standard.

Noninvasive detection of cardiovascular pulsations by optical Doppler techniques

A system has been developed based on the measurement of skin surface vibration that can be used to detect the underlying vascular wall motion of superficial arteries and the chest wall. Data obtained from tissue phantoms suggested that the detected signals were related to intravascular pressure, an important clinical and physiological parameter. Unlike the conventional optical Doppler techniques that have been used to measure blood perfusion in skin layers and blood flow within superficial arteries, the present system was optimized to pick up skin vibrations. An optical interferometer with a 633-nm He:Ne laser was utilized to detect micrometer displacements of the skin surface. Motion velocity profiles of the skin surface near each superficial artery and auscultation points on a chest for the two heart valve sounds exhibited distinctive profiles. The theoretical and experimental results demonstrated that the system detected the velocity of skin movement, which is related to the time derivative of the pressure. The system also reduces the loading effect on the pulsation signals and heart sounds produced by the conventional piezoelectric vibration sensors. The system's sensitivity, which could be optimized further, was 366.2 mm/s for the present research. Overall, optical cardiovascular vibrometry has the potential to become a simple noninvasive approach to cardiovascular screening.

Noninvasive detection of cardiovascular pulsations by optical Doppler techniques

A system has been developed based on the measurement of skin surface vibration that can be used to detect the underlying vascular wall motion of superficial arteries and the chest wall. Data obtained from tissue phantoms suggested that the detected signals were related to intravascular pressure, an important clinical and physiological parameter. Unlike the conventional optical Doppler techniques that have been used to measure blood perfusion in skin layers and blood flow within superficial arteries, the present system was optimized to pick up skin vibrations. An optical interferometer with a 633-nm He:Ne laser was utilized to detect micrometer displacements of the skin surface. Motion velocity profiles of the skin surface near each superficial artery and auscultation points on a chest for the two heart valve sounds exhibited distinctive profiles. The theoretical and experimental results demonstrated that the system detected the velocity of skin movement, which is related to the time derivative of the pressure. The system also reduces the loading effect on the pulsation signals and heart sounds produced by the conventional piezoelectric vibration sensors. The system’s sensitivity, which could be optimized further, was 366.2 mm/s for the present research. Overall, optical cardiovascular vibrometry has the potential to become a simple noninvasive approach to cardiovascular screening. © 1997 Society of Photo-Optical Instrumentation Engineers. [S1083-3668(97)00604-7]

Automated phonocardiographic screening for heart disease in children.

Mass screening for heart disease in children can be accomplished by automatic analysis of phonocardiograms (PCG) obtained from five routine auscultation points. An ECG signal serves as reference. PCG signals exceeding 20% of the amplitude of the first heart sound S1 in the diastolic interval and 40% of S1 in the systolic interval are considered as abnormal. Signals in the systolic interval with an amplitude between 20 and 40% of S1 are classified as normal if their duration is less than two-thirds of the interval from the first to second heart sound. Thus, the method takes into account innocent murmurs and classifies them as normal. The splitting of the second sound is detected by monitoring its duration at a level of 20% of S1. This method was tested on 2,583 children aged from 3 to 14 years and on 345 children with proven heart disease. The results obtained showed a sensitivity of 96.5% and a specificity of 92.4%.

Auricular heart sounds in auricular flutter

1. 1. Two cases of auricular flutter in which audible auricular sounds were graphically recorded are added to the eight cases already reported. 2. 2. The cases presented were unusual in the following respects: 2.1. a. The exceptional loudness of the auricular sounds, which approached or exceeded the ventricular sounds in intensity, 2.2. b. The point of maximal intensity in the pulmonic and aortic auscultation points, respectively, 2.3. c. The presence of palpable and mechanically recordable impulses synchronous with the auricular sounds and the flutter waves in one case, 2.4. d. The intensification of the sounds by the intravenous injection of a digitalis glycoside in the other case, 2.5. e. Perfect synchrony with the flutter wave of the electrocardiogram except in the cycle immediately following ventricular systole, where there was a slight but significant delay in the appearance of the sound, 2.6. f. The clicking character of the sounds, 2.7. g. The occurrence of a loud auricular sound during ventricular systole. 3. 3. Two alternate explanations are offered for the sounds observed: The flutter contractions of the auricles may have been audible because of their unusual forcefulness, or an “auricular click” associated with pleuropericardial adhesions may have been present.