Explore chapters and articles related to this topic
Application of Soft Computing Techniques to Heart Sound Classification
Published in Ashish Mishra, G. Suseendran, Trung-Nghia Phung, Soft Computing Applications and Techniques in Healthcare, 2020
Traditionally, cardiologists use stethoscopes for examination of heart sounds. The accuracy of heart sound classification is based on the experience and skill of the physicians. But this manual clinical process is time-consuming and costly. To alleviate these limitations, recently a computer-based automatic computer assist tool is recommended for detection of abnormal heart sound. Hence this is becoming an emerging research for the biological signal processing and machine learning groups as it is computer based. Soft computing is one of the problem-solving approaches used to solve real life complex problems in the field of science and technology. Applications of various soft computing techniques such as artificial neural network, fuzzy logic and evolutionary computing have been extensively used in the medical diagnosis. Various soft computing techniques are also applied by the researchers in the field of classification of heart sound.
Biomedical Sensors and Data Acquisition
Published in Rajarshi Gupta, Dwaipayan Biswas, Health Monitoring Systems, 2019
Each cardiac cycle is considered to comprise of two periods, viz., systole, representing the contraction or depolarization of cardiac chambers, and diastole, representing the relaxation or repolarization of cardiac chambers. Generation of ECG has already been briefly stated in Section 2.3.1. The PCG consists of four major components I, II, III, and IV, also named as S1, S2, S3, and S4. Out of these, S1 and S2 are generated due to valve closure, have the largest intensity, and are audible as ‘lub’ and ‘dub’, respectively. The first heart sound S1 is observed between closing of mitral (and bicuspid) valve and opening of aortic (and pulmonary) valve. The second heart sound (S2) is audible between the exactly opposite events, i.e., opening of aortic (and pulmonary) valves and closing of tricuspid (and bicuspid) valves. S3 and S4 are with comparatively dull and weak in intensity, observed in children and certain adults, and not related to valve activity.
Other Biomedical Signals
Published in Kayvan Najarian, Robert Splinter, Biomedical Signal and Image Processing, 2016
Kayvan Najarian, Robert Splinter
A signal used for diagnostics of the heart is the heart sound. The heart sounds are the sounds made by the flow of the blood in and out of the heart compartments. In order to measure the heart sounds, often mechanical stethoscopes are used to amplify the sounds. However, since these devices are known to have an uneven frequency response, they somehow distort the sounds. From the signal processing point of view, these changes in the heart sounds made by the mechanical stethoscopes are direct filtering of the actual sounds (i.e., inputs) based on the internal structure of the stethoscope (i.e., mechanical filter), which provides an altered perceived sound (output). While electronic stethoscopes overcome these problems and provide much less distorted version of the actual sounds, physicians have not generally accepted these electronic devices.
Towards classifying non-segmented heart sound records using instantaneous frequency based features
Published in Journal of Medical Engineering & Technology, 2019
Heart sounds represent the basic element for evaluating a patient’s cardiovascular function. Cardiovascular functionality can be assessed using biosignals such as Electrocardiogram (ECG) and Phonocardiogram (PCG), then a set of features are extracted from these biosignals and fed into machine learning algorithms to classify diseases heart valves [3,4]. Because it is easiest to record the heart sound than recording the electrical activity of the heart, in addition to its required less wiring connection, PCG signals take the researcher's attention to exploit it as a significant data source about the health status of the heart. The PCG signal pattern differs with respect to heart disease, especially which are related to valve functionality. A big difference in the pattern and the shape can be noticed between the normal and abnormal heart sound as their signal varies from disease to other with respect to time, amplitude, intensity, and frequency content [5,6].
Wavelet bispectrum-based nonlinear features for cardiac murmur identification
Published in Cogent Engineering, 2018
Dinesh Kumar, Rajendrasinh Jadeja, Sarang Pande
In shifting paradigm of healthcare system from curative to preventive, it has been inevitable to capture health conditions via monitoring system over a significant time in order to avoid critical situation like surgery. In this attempt, personalized healthcare devices are being developed which enable acquiring of physiological signals and their analysis. Furthermore, with these systems patients can perform follow-up themselves. Heart sound is one of the potential and economic physiological signals which can be used to determine heart’s conditions, namely in the cases of valvular dysfunction. Therefore, efficient analysis techniques are required to process heart sound and extract relevant information related to such diseases. Abnormal sounds are heard besides the regular heart beating sound in cases of heart-related problems. This paper proposes a new technique for abnormal heart sound analysis whereby presence of certain cardiac diseases can automatically be determined.
A hybrid algorithm for heart sounds segmentation based on phonocardiogram
Published in Journal of Medical Engineering & Technology, 2019
After hearing the first sound, blood enters the arteries. The second heart sound is sharper but shorter than S1. The S2 is heard during ventricles rest due to the closure of the semilunar valves. After hearing the S2, the blood enters the ventricles. The distance between S1 and S2 corresponds to the time of the heart systole and this time is naturally shorter than the time between S2 and S1, which corresponds to diastole. In physiological conditions, there is silence at systole and diastolic distances. S1 has a duration of 70–150 ms, and S2 has a length of 60–120 ms [6]. The sound S1 has a frequency range of 20–150 Hz and S2 a frequency range of 50–250 Hz [7].