Researchers Develop Computer-Aided Classification of Heart Sounds Not Detectable to the Human Ear
July 17, 2008 - Heart sounds provide valuable diagnostic information concerning the heart valves and hemodynamics of heart, but the poor sensitivity of human ears in the low frequency range makes this task difficult, so two researchers have created a computer-aided phonocardiogram (PCG) to analyze a digitized recordings of heart sounds.
Certain heart diseases are best detected only by auscultation. The collection of sonic waves from the surface of the body, as performed with the stethoscope, continues to provide an important source of clinical information that, together with the overall bedside examination, is cost-effective and also cannot be totally replaced by alternative technical methods like echocardiography. Moreover, echocardiography is not required for all patients with systolic murmurs. However, it is reported that a disturbing percentage of medical graduates cannot properly diagnose heart conditions using stethoscope.
For this reason, researchers Samit Ari and Goutam Saha of the at Indian Institute of Technology’s Department of Electronics and Electrical Communication Engineering in Kharagpur, India developed a computer-aided stethoscope to more accurately analyze heart sounds. Their findings are explained in the inaugural issue of the International Journal of Medical Engineering and Informatics in published in July.
Their system extracts robust acoustic features for automatic classification of heart sounds based on empirical mode decomposition (EMD). The work decomposes segmented heart sound cycles with EMD to generate certain intrinsic mode functions (IMF). The first IMF contains mostly high frequency noise, the second and third IMFs carry higher frequency components of signals of interest and the residue contains its low frequency components. The system examines 25 dimensional feature vectors,
They conducted experiments on 104 different recordings of heart sound comprising of normal and 12 different pathological cases against three different additive background noises - white Gaussian, hospital and body noise. It is found that the EMD based feature extraction always performed better than benchmark wavelet based feature extraction technique.
The acoustic energy produced by the mechanical activity of the heart can produce tell-tale sounds of abnormalities in cardiac components and manifest itself in the corresponding sounds in the PCG. Unlike echocardiography, the non-invasive cardiac auscultation is simple, cost-effective and with proper signal processing and pattern recognition tools may emerge as an important device for primary detection of heart valve disorders, the researchers said.
A cardiac cycle primarily contains two major sounds – first heart sound (S1) followed by second heart sound (S2). The intervals between S1 and S2 (systole), and S2 and S1 (diastole) of the next cycle are usually silent for normal cases. These two distinct normal heart sounds are often described as lub and dub (or dup), and occur in sequence with each heartbeat. In an abnormal heart sounds there could be several other sounds in the PCG signal besides primary heart sounds. Abnormal heart murmurs, if present in the cardiac cycle, refer to different pathological conditions as per location, shape, duration and other associated features. Murmurs are generally high frequency, noise-like sounds that arise when the velocity of blood become high when it flows through an irregularity. The PCG signal has been widely used to detect the cardiac abnormalities. Different features of PCG signals like intensity, frequency content, split information, time relations, etc. can give an idea of the underlying pathology, if any, and the state of the heart function.
For more information: www.inderscience.com
More like this
- Researchers Develop Computer-Aided Classification of Heart Sounds Not Detectable to the Human Ear
- Accuro Diagnostic Imaging
- Accuro Diagnostic Imaging
- R2 CAD Features Sophisticated Pattern Recognition Software
- Automatic Deformable Alignment Significantly Reduces Radiologist Time to Match Lung Nodule Locations