Phonocardiographic Signal Analysis Method Using a Modified Hidden Markov Model

We describe the development of an algorithm for the automatic classification of heart sound phonocardiogram waveforms as normal, abnormal or uncertain. Our approach consists of three major components: 1) Heart sound segmentation, 2) Transformation of one-dimensional waveforms into two-dimensional timefrequency heat map representations using Mel-frequency cepstral coefficients and 3) Classification of MFCC heat maps using deep convolutional neural networks. We applied the above approach to produce submissions for the 2016 PhysioNet Computing in Cardiology Challenge. We present results from the challenge, as well as describe in detail the resulting neural network architecture produced and design decisions made. IntroductionThe goal of the 2016 PhysioNet Computing in Cardiology Challenge was to accurately classify normal and abnormal heart sounds from phonocardiogram (PCG) waveforms. A particular aim was to identify from a single short recording whether a subject should be referred on for expert diagnosis. Accurate and robust algorithms were required that could deal with heart sounds that exhibit very poor signal quality.The challenge training set consisted of 3,240 heart sound recordings, lasting from 5 seconds to just over 120 seconds. Recordings were collected from nine different locations on the body (including aortic area, pulmonic area, tricuspid area and mitral area, among others). Recordings from healthy subjects were labeled as normal. Recordings from subjects with a confirmed cardiac diagnosis were labeled as abnormal. Abnormal recordings were collected from patients who suffered from a variety of illnesses, including heart valve defects (mitral valve prolapse, mitral regurgitation, aortic stenosis, valvular surgery) and coronary artery disease. An in depth description of the challenge and the dataset is provided in [1], as well as a thorough review of the field of cardiac auscultation.We present an algorithm that computes heat maps of the time-frequency distribution of signal energy and uses a deep convolutional neural network to automatically classify normal versus abnormal heart sound recordings. Logistic regression hidden semi-Markov model-based heart sound segmentation is first performed on the PCG waveform. Spectrograms (energy maps) consisting of 6 cepstral coefficients that capture Mel-frequencies varying over time are derived for overlapping sliding windows of threesecond duration, beginning at the first heart sound, S 1 . A deep convolutional neural network consisting of two alternating convolution and max pooling layers is trained to perform automatic feature extraction. A final multilayer perceptron, consisting of two fully connected layers, distinguishes between normal and abnormal spectrograms. Heart sound recordings of variable length are dealt with by computing an ensemble of logit scores for all overlapping three-second segments contained within a recording and maximizing the average scores computed for all classes. ApproachOur approach consists of three major components: 1. Segme...

“…MFCC are commonly used as a feature type in automatic speech recognition [4]. Previous approaches to cardiac auscultation have also utilized MFCCs [5].…”

Section: Mfccmentioning

confidence: 99%

Classifying Heart Sound Recordings using Deep Convolutional Neural Networks and Mel:Frequency Cepstral Coefficients

Rubin¹,

Abreu²,

Ganguli³

et al. 2016

“…Tamer et al proposed a wavelet-based segmentation and artificial neural network based classification [6]. Guy et al proposed a clustering based approach in [7] [9,10]. Despite the availability of literature, there is still paucity on the sensitivity of the study due to lack of a standardized and high-quality database of heart sound recordings.…”

Section: Introductionmentioning

confidence: 99%

Heart Sound Classification from Wavelet Decomposed Signal Using Morphological and Statistical Features

Munia

Tavakolian

Verma

et al. 2016

“…These segmentation techniques use different approaches such as signal envelopes With the segmented cardiac cycles, the classification of heart sound pathologies is made possible and several methods have been proposed over the last decades. Among these studies, artificial neural networks [13], support vector machines [14] and HMM based [15] approaches are common. Classification based on clustering has also been shown to be effective in heart sound pathology classification [16].…”

Section: Introductionmentioning

confidence: 99%

“…Among these studies, artificial neural networks [13], support vector machines [14] and HMM based [15] approaches are common. Classification based on clustering has also been shown to be effective in heart sound pathology classification [16].…”

Section: Introductionmentioning

confidence: 99%

Identification of Abnormal Heart Sounds

Yazdani

Schlatter

Atyabi

et al. 2016

As a part of 2016 Physionet/CinC callenge, this work aims at the detection of abnormal phonocardiogram (PCG) recordings. Heart sound signal analysis has been an active research topic over the past decades with various studies such as heart sound segmentation and classification. We used the Physionet/CinC2016 challenge PCG database, which contains a large public collection of PCG recordings from a variety of clinical and nonclinical environments. The PCG classification in this work is performed in two steps. PCG heartbeats are first segmented and various heart sound markers are delineated. Then, a series of beat-specific features are extracted from the segmented heartbeats. Finally, PCG recordings are classified into normal and abnormal groups by performing classification based on tape-long features and by analyzing beatextracted features from the PCG. Our method achieved an overall score of 80 in the unofficial phase of the challenge. In the official phase, the overall score of the proposed method was 82, with a sensitivity of 89%. IntroductionMechanical activity of the heart, which is triggered by the propagation of electrical activity within the heart, is audible at different locations on the chest wall. Phonocardiogram (PCG) is an audio recording of these mechanical activities recorded at the chest surface. Heat sounds provide valuable information which can help in the diagnosis of heart valve disorders. PCG comprises fundamental heart sounds from which S1 and S2 are the most prominent in normal recordings, respectively representing the beginning of ventricular contraction and the beginning of diastole [1]. In abnormal cases, various markers maybe present beside the fundamental heart sounds such as murmurs. Murmurs are noise-like high frequency sounds and long systolic murmurs as well as diastolic and continuous murmurs are generally pathologic [2].Auscultation is a common cost-effective technique that provides valuable information about heart valve problems. However, by listening to mechanical activity of the heart with a stethoscope, physicians cannot examine all physical characteristics of heart sounds, as it has been shown that only limited activity of heart sounds and murmurs are within human audibility range [1] [3].Over the years, various heart sound classification methods have been proposed. Generally, the analysis of heart sounds is performed on the heart cycle. To this end many PCG heart sound segmentation methods have been developed, enabling the detection of fundamental heart sound markers such as the beginning/end of S1, systole, S2 and diastole. These segmentation techniques use different approaches such as signal envelopes With the segmented cardiac cycles, the classification of heart sound pathologies is made possible and several methods have been proposed over the last decades. Among these studies, artificial neural networks [13], support vector machines [14] and HMM based [15] approaches are common. Classification based on clustering has also been shown to be effective in heart sound...