Feature Extraction for Systolic Heart Murmur Classification

Abstract

Abstract-Heart murmurs are often the first signs of pathological changes of the heart valves, and they are usually found during auscultation in the primary health care. Distinguishing a pathological murmur from a physiological murmur is however difficult, why an ''intelligent stethoscope'' with decision support abilities would be of great value. Phonocardiographic signals were acquired from 36 patients with aortic valve stenosis, mitral insufficiency or physiological murmurs, and the data were analyzed with the aim to find a suitable feature subset for automatic classification of heart murmurs. Techniques such as Shannon energy, wavelets, fractal dimensions and recurrence quantification analysis were used to extract 207 features. 157 of these features have not previously been used in heart murmur classification. A multi-domain subset consisting of 14, both old and new, features was derived using PudilÕs sequential floating forward selection (SFFS) method. This subset was compared with several single domain feature sets. Using neural network classification, the selected multi-domain subset gave the best results; 86% correct classifications compared to 68% for the first runner-up. In conclusion, the derived feature set was superior to the comparative sets, and seems rather robust to noisy data.

A robust heart sounds segmentation module based on S-transform

“…In [23], authors extended the method to also incorporate information from the eigenvectors to classify EEG seizures. In [24], the last technique is applied on the S-matrix in the aim to extract features for systolic heart murmur classification. Following this approach, this study proposes a feature extraction method for S1 and S2 classification.…”

Diagnosis of Heart Valve Disorders through Trapezoidal Features and Hybrid Classifier

“…Most of these researches were dedicated to distinguish normal heart sounds from murmurs and the rest were tried to distinguish different types of murmurs. Ahlstrom et al [9] proposed different linear and non-linear features based on DWT, and classified a set of systolic murmurs with 86% accuracy. In another work by Ahlstrom et al [15], recurrence quantification analysis was proposed to differentiate aortic stenosis from innocent murmurs, and 90%sensitivity and 88% specificity were achieved.…”

“…A number of features proposed based on wavelet transform to Diagnosis of Heart Valve Disorders through Trapezoidal Features and Hybrid Classifier Fatemeh Safara, Shyamala Doraisamy, Azreen Azman, Azrul Jantan, and Sri Ranga differentiate murmurs from normal heart sounds. Ahlstrom et al [9], applied techniques such as entropy, energy, recurrence quantification analysis and Fractal dimension to generate features, and classify systolic murmurs. Choi et al [10] defined two features, meanWPE and stdWPE, based on wavelet packet energy (WPE); and the contribution of these two types of features was evaluated using accuracy achieved by support vector machine (SVM) classifier.…”

Phonocardiogram Signal Processing Module for Auto-Diagnosis and Telemedicine Applications

“…In [20] authors extended the method to also incorporate information from the eigenvectors to classify EEG seizures. In [21] the last technique is applied on the S-matrix in the aim to extract features for systolic heart murmur classification. Following this approach, this study proposes a feature extraction method for S1 and S2 classification.…”