Heart sound model based on DenseNet121 architecture for diagnosis of aortic stenosis: A prospective clinical trial / 中国胸心血管外科临床杂志

@#Objective     To identify the heart sounds of aortic stenosis by deep learning model based on DenseNet121 architecture, and to explore its application potential in clinical screening aortic stenosis. Methods      We prospectively collected heart sounds and clinical data of  patients with aortic stenosis in Tianjin Chest Hospital, from June 2021 to February 2022. The collected heart sound data were used to train, verify and test a deep learning model. We evaluated the performance of the model by drawing receiver operating characteristic curve and precision-recall curve.  Results     A total of 100 patients including 11 asymptomatic patients were included. There were 50 aortic stenosis patients with 30 males and 20 females at an average age of 68.18±10.63 years in an aortic stenosis group (stenosis group). And 50 patients without aortic valve disease were in a negative group, including 26 males and 24 females at an average age of 45.98±12.51 years. The model had an excellent ability to distinguish heart sound data collected from patients with aortic stenosis in clinical settings: accuracy at 91.67%, sensitivity at 90.00%, specificity at 92.50%, and area under receiver operating characteristic curve was 0.917.   Conclusion     The model of heart sound diagnosis of aortic stenosis based on deep learning has excellent application prospects in clinical screening, which can provide a new idea for the early identification of patients with aortic stenosis.

Heart sound classification algorithm based on time-frequency combination feature and adaptive fuzzy neural network / 生物医学工程学杂志

Feature extraction methods and classifier selection are two critical steps in heart sound classification. To capture the pathological features of heart sound signals, this paper introduces a feature extraction method that combines mel-frequency cepstral coefficients (MFCC) and power spectral density (PSD). Unlike conventional classifiers, the adaptive neuro-fuzzy inference system (ANFIS) was chosen as the classifier for this study. In terms of experimental design, we compared different PSDs across various time intervals and frequency ranges, selecting the characteristics with the most effective classification outcomes. We compared four statistical properties, including mean PSD, standard deviation PSD, variance PSD, and median PSD. Through experimental comparisons, we found that combining the features of median PSD and MFCC with heart sound systolic period of 100-300 Hz yielded the best results. The accuracy, precision, sensitivity, specificity, and F1 score were determined to be 96.50%, 99.27%, 93.35%, 99.60%, and 96.35%, respectively. These results demonstrate the algorithm's significant potential for aiding in the diagnosis of congenital heart disease.

Automatic heart activity diagnosis based on Gram polynomials and probabilistic neural networks

The paper proposes a new approach to heart activity diagnosis based on Gram polynomials and probabilistic neural networks (PNN). Heart disease recognition is based on the analysis of phonocardiogram (PCG) digital sequences. The PNN provides a powerful tool for proper classification of the input data set. The novelty of the proposed approach lies in a powerful feature extraction based on Gram polynomials and the Fourier transform. The proposed system presents good performance obtaining overall sensitivity of 93%, specificity of 91% and accuracy of 94%, using a public database of over 3000 heart beat sound recordings, classified as normal and abnormal heart sounds. Thus, it can be concluded that Gram polynomials and PNN prove to be a very efficient technique using the PCG signal for characterizing heart diseases.

El corazón y sus ruidos cardíacos normales y agregados. Una somera revisión del tema / The heart, normal sounds and murmurs

Resumen Los ruidos cardíacos son la expresión sonora del cierre de las válvulas cardíacas, su funcionamiento fisiológico siempre es unidireccional, lo cual permite la correcta circulación de la sangre a través del circuito cardiovascular. La auscultación del área precordial permite la identificación de estos ruidos y sus matices en los 5 focos de auscultación. Existen ruidos que no son producidos por el cierre de las válvulas, por mencionar algunos podemos encontrar los llamados soplos y los ruidos de Korotkoff, ambos producidos por la interrupción del flujo natural de la sangre (flujo laminar) al convertirse en flujo turbulento cada vez que se encuentra una disminución del radio de los conductos por donde ésta circula.

Abstract The heart sounds are an audible expression of the heart valves closing. Their physiological function is always unidirectional, allowing the proper blood flow through the cardiovascular circuit. Listening - by auscultation- to the specific chest areas allows the identification of these sounds and nuances in the five auscultation areas. There are sounds that are not produced by the closing of the valves; to mention a few, we can find the so-called puffs and Korotkoff sounds, both produced by interrupting the natural flow of blood flow (laminar flow) that becomes a turbulent flow whenever there is a reduction of the radius of the conduits through which this the blood circulates.

Signal Processing Tools for Heart Sounds Analysis Based on Time-Frequency Domain.

This paper present several signal processing tools for the analysis of heart sounds. Cardiac auscultation is noninvasive, low-cost and accurate to diagnose some heart diseases. A new module for the segmentation of heart sounds based on S-Transform is presented. The heart sound segmentation process divides the Phono Cardio Gram (PCG) signal into four parts: S1 (first heart sound), systole, S2 (second heart sound) and diastole. The segmentation can be considered one of the most important phases in the auto-analysis of PCG signals. A segmentation method based on the Shannon energy of the local spectrum calculated by the S-transform is proposed. Then, the energy concentration of the S-transform is optimized to accurately detect the boundaries of the localized sounds. New features based on the energy concentration of the S-transform are proposed to classify S1 and S2 and other features based on the complexity measu Frequency (TF) domain are proposed to detect systolic murmurs.

Modelos Acústicos HMM Multimodales para Sonidos Cardiacos y Pulmonares / Multimodal acoustic models HMM applied to heart and lung sounds

Este artículo muestra el proceso de clasificación de señales bioacústicas normales y anormales registradas sobre el tórax humano lo cual incluye los sonidos de corazón y del pulmón. La idea específica es diseñar un sistema de clasificación de señales basado en técnicas de modelado acústico empleando particularmente modelos HMM para detectar secuencias de eventos, y GMM para modelar cúmulos que corresponden a los datos de los eventos. Las modalidades para extraer las características de los datos son vectores MFCC y Octiles. Esta aproximación tiene el potencial de mejorar la clasificación de la precisión en indicadores de diagnóstico auscultatorios, esto es interesante ya que los modelos HMM han demostrado ser menos sensibles al ruido en estudios previos. Resultados preliminares demuestran una precisión del 95% en clasificación de las señales de sonido evaluadas. Esto es particularmente critico tomando en cuenta la interferencia ambiental en una variedad de consultorios médicos. Debido a que algunas frecuencias del sonido cardiaco son paralelas a los sonidos pulmonares, estas pueden ser modeladas a partir de un mismo registro. Resultados experimentales preliminares de esta aproximación demuestran que es factible el desarrollo de valoraciones de diagnóstico automatizado de pacientes mediante identificadores de diagnóstico auscultatorios en forma temprana usando tecnologías de bajo costo.

This paper demonstrates classification processes of normal and abnormal bioacoustics signals recorded over a human thorax which encompasses heart and lung sounds. The specific aim is to design a signal classification system based on acoustical modeling techniques employing particularly HMM models to detect events' sequences, and GMM to model clusters corresponding to the data events. The modalities for extracting data characteristic are the MFCC and Octile vectors. These approaches have a potential of enhancing the classification accuracy of these auscultatory diagnostic indicators as the initial studies demonstrated that the HMM based models are less sensitive to the noise. Preliminary results demonstrate over 95% accuracy in classification of the evaluated sound signals. This is particularly critical taking into account environmental interference in a variety of medical care settings. As the heart sounds frequency components parallel those of the lungs sounds, but with a different periodicity, they can be modeled with the same recording. The preliminary experimental results are supportive of this approach and demonstrate feasibility of a development of an automated early diagnostic assessment of patients' auscultatory diagnostic indicators utilizing low cost technologies.

Heart Sound Classification Using Variable Number of States in Hidden Markov Model Considering Characteristics of the Signal / 대한의료정보학회지

Hidden Markov model (HMM) is known to be one of the most powerful methods in the acoustic modeling of heart sound signals. Conventionally, we usually use a fixed number of states for each HMM. However, due to the various types of the heart sound signals, it seems that more accurate acoustic modeling is possible by varying the number of states in the HMM depending on the signal types to be modeled. In this paper, we propose to assign different number of states to the HMM for better acoustic modeling and consequently, improving the classification performance of the heart sound signals. Compared with when fixing the number of states, the proposed approach has shown some performance improvement in the classification experiments on various types of heart sound signals.

Classification of Continuous Heart Sound Signals Using the Ergodic Hidden Markov Model / 대한의료정보학회지

Recently, hidden Markov models (HMMs) have been found to be very effective in classifying heart sound signals. For the classification based on the HMM, the continuous cyclic heart sound signal needs to be manually segmented to obtain isolated cycles of the signal. However, the manual segmentation will be practically inadequate in real environments. Although, there have been some research efforts for the automatic segmentation, the segmentation errors seem to be inevitable and will result in performance degradation in the classification. To solve the problem of the segmentation, we propose to use the ergodic HMM for the classification of the continuous heart sound signal. In the classification experiments, the proposed method performed successfully with an accuracy of about 99(%) requiring no segmentation information.

Usefulness of the Time-Axis Variable Function of the Digital Stethoscope in Teaching Cardiac Auscultation / 医学教育

The purpose of this study was to assess the usefulness of the time-axis variable function of the digital stethoscope in teaching cardiac auscultation. The subjects were 61 fifth-year medical students. The effectiveness of the time-axis variable function of the digital stethoscope for cardiac auscultation was assessed with five representative heart sounds or murmurs. The students reported that auscultatory findings were clearer at half speed than at normal speed for the following sounds, in decreasing order of frequency: systolic murmur following a systolic click, splitting of the second heart sound, systolic ejection murmur, and the third heart sound. In contrast, students frequently reported that auscultatory findings were clearer at normal speed for systolic regurgitation murmur. We suggest that the time-axis variable function is useful for improving auscultatory ability and would be particularly effective for helping students identify and understand the temporal relation of sequential heart sounds or of multicomponent murmurs.

Development of heart sounds teletransmission system based on GPRS / 医疗卫生装备

Heart sounds teletransmission is very important in telemedicine.The heart sounds are transmitted to the server which has been accessed to internet in custody center based on GPRS,then doctors analyze the heart sounds and transmit the conclusion to the consumer.

The Heart Sound Data Analysis, Management System using the Wireless Stethoscope / 대한의료정보학회지

This paper is about the system using a wireless stethoscope to analysis the FFT and the time-frequency for a heart sound and to manage the collected data for a web-based system. We reformed a wireless stethoscope, connected to PC interface and added the analysis function. In result, we combined merits of an existed wireless system to be convenient for measuring the heart sound and to be available for many listener to ausculate the heart sound simultaneously, and an existed wired system to supply the various analysis functions. The heart sounds data was made into the database to search or refer to the patient data. It is possible to search and refer by the web-browser to the recorded heart sound file, the analyzed file by FFT and the STFT time-frequency method.