CHAPTER 4 CLASSIFICATION OF HEART MURMURS USING WAVELETS AND NEURAL NETWORKS

Size: px
Start display at page:

Download "CHAPTER 4 CLASSIFICATION OF HEART MURMURS USING WAVELETS AND NEURAL NETWORKS"

Transcription

1 52 CHAPTER 4 CLASSIFICATION OF HEART MURMURS USING WAVELETS AND NEURAL NETWORKS 4.1 INTRODUCTION Heart auscultation is the process of interpreting sounds produced by the turbulent flow of blood into and out of the heart and the movement of mechanical structures that control this flow. This non-invasive, low-cost screening technique is used as a primary tool in the diagnosis of certain heart disorders, especially valvular problems. The conventional method of auscultation with a stethoscope has many limitations. The skills required for interpretation are acquired by listening to the heart sounds of many different patients. Therefore, it is a very subjective process that depends largely on the physician s experience and ability to differentiate between different sound patterns. For an objective assessment of heart sounds for diagnosis of certain cardiac disorders, digital recording and subsequent analysis of these acoustic signals is the only reliable method. Digital sounds can be analysed efficiently using a computer. 4.2 LITERATURE SURVEY Faizan et al (2006) have developed a heart murmur classification system with features extracted using Spectrogram. Seven types of murmurs are classified using a Multilayer Perceptron Neural Network based on their timing within the cardiac cycle using Smoothed Pseudo Wigner-Ville distribution. Zhou et al (2008) have proposed a heart sound recognition

2 53 system based on features extracted from the normalized average Shannon energy in wavelet domain. These features are used with Full Bayesian Neural Networks to classify six types of heart sounds. Yadi et al (2008) report a classification method based on statistical features extracted using wavelet decomposition and MLP Neural Network to classify five murmurs. Many new techniques have been introduced for efficient heart disease diagnosis. However, they are expensive, require skilled technicians to operate the equipment and experienced cardiologists to interpret the results (Faizan et al 2006). These facilities are usually available only in advanced hospitals and not in rural and most urban hospitals. Furthermore, the wavelet transform has demonstrated the ability to analyse the heart murmurs more accurately than the techniques like STFT or Wigner Ville Distribution (WVD) cited in literature. The WVD has the ability to separate signals in both time and frequency. One advantage of the WVD over the STFT is that it does not suffer from the time frequency trade-off problem. On the other hand, WVD has a disadvantage since it shows cross terms in its response and attempts to smooth these results in decreased resolution in both time and frequency (Debbal and Bereksi-Reguig 2008). Considering these constraints, a new approach based on wavelet transform and artificial neural networks has been proposed for classifying heart murmurs into eight types, namely, normal, early systolic, mid systolic, late systolic, holo systolic, early diastolic, mid diastolic and late diastolic. Classification of heart murmurs into more number of types leads to enhanced accuracy of diagnosis. 4.3 HEART MURMURS Categories In healthy adults, there are two normal heart sounds often described as a lub and a dub, that occur in sequence with each heart beat. These are the first heart sound (S1) and second heart sound (S2), produced by the turbulent flow against the closed atrioventricular and semilunar valves respectively.

3 54 The period between S1 and S2 is referred to as systole and the period between S2 and subsequent S1 is diastole. Murmurs are abnormal sounds that occur between S1 and S2 called systolic murmurs and between S2 and S1 called diastolic murmurs. The former occurs during ventricular contraction (systole) and the latter occur during ventricular filling (diastole). These heart sounds and murmurs are shown in Figure 4.1. (a) Normal heart sounds S1 and S2 (b) Systolic murmurs between S1 and S2 (c) Diastolic murmurs between S2 and S1 Figure 4.1 Heart sounds and murmurs

4 55 The systolic murmurs are again sub classified into early, mid, late and holo systolic murmurs depending on their position in the systolic period. Similarly the diastolic murmurs are further sub classified into early, mid and late diastolic murmurs. These pathological types of heart murmurs and their associated abnormalities are shown in Table 4.1. Table 4.1 Heart murmurs and associated abnormalities Category Sub category Associated abnormality Systolic Diastolic Early systolic Mid systolic Late systolic Holo systolic Early diastolic Mid diastolic Late diastolic Mitral regurgitation Aortic stenosis Atrial septal defect Mitral valve prolapse Tricuspid valve prolapse Papillary muscle dysfunction Tricuspid insufficiency Ventricular septal defect Aortic regurgitation Pulmonary regurgitation Left anterior descending artery stenosis Mitral stenosis Tricuspid stenosis Atrial myxoma Complete heart block Characteristics Murmurs can be classified based on seven different characteristics : timing, shape, location, radiation, intensity, pitch and quality. Timing refers to whether the murmur is a systolic or diastolic murmur. Shape refers to the intensity over time; murmurs can be crescendo, decrescendo or crescendodecrescendo. Location refers to where the heart murmur is auscultated best. There are six places on the anterior chest to listen for heart murmurs; each of these locations roughly corresponds to a specific part of the heart. Radiation

5 56 refers to where the sound of the murmur radiates. The general rule of thumb is that the sound radiates in the direction of the blood flow. Intensity refers to the loudness of the murmur, and is graded on a scale from 0-6/6. The pitch of the murmur is low, medium or high and is determined by whether it can be auscultated best with the bell or diaphragm of a stethoscope. The quality of a murmur may be blowing, harsh, rumbling and musical. In the proposed work, classification is done with respect to the timing characteristic. 4.4 HEART MURMUR CLASSIFICATION Data Collection and Pre-Processing Heart murmur files, normal and pathological, collected from the Johns Hopkins Cardiac Auscultatory Recording Database (CARD), an excellent resource for research purposes, and other internet sites, were used in this classification work (CARD 2006; Raymond 2006). The pathological types include early systolic, mid systolic, late systolic, holo systolic, early diastolic, mid diastolic and late diastolic murmurs. The murmur files obtained in mp3 format were converted to wav format for further processing. The sampling frequency chosen was 8000Hz in all cases to cover the range for analysing murmurs (Zhou et al 2008). Signals, when contaminated with noise, might lead to erroneous classification. To remove the noise that may overlap with the heart sounds, SureShrink, the Discrete Wavelet Transform (DWT) based wavelet shrinkage denoising technique was used (Donoho and Johnstone 1995; Cai and Harrington 1998) Feature Extraction The main objective of feature extraction process is to derive a set of features that best represents the signal. So, the selection of features is an important criterion for proper classification of different heart murmurs. In this work, a set of 15 statistical features were extracted from the murmur files. The steps involved in the feature extraction process are illustrated in Figure 4.2.

6 57 Digitized heart sound sampled at 8000 Hz 16 bits/sample Extraction of single cycle of the heart murmur Normalized to absolute maximum Wavelet decomposition (db6 level 5) 3 rd level detail Hz 4th level detail Hz 5th level detail Hz 5th level approximation Hz Determining mean of absolute values of coefficients in each sub band Determining standard deviation of wavelet coefficients in each sub band Determining average power of wavelet coefficients in each sub band Determining ratio of absolute mean values of adjacent sub bands Figure 4.2 Feature extraction process

7 58 The denoised signal was segmented into cardiac cycles. A single cycle containing first heart sound S1, systolic period, second heart sound S2 and diastolic period, was extracted for analysis. To nullify the effect of input gain variations, the original signal was normalized to absolute maximum. Heart murmurs are non-stationary signals exhibiting marked changes with time and frequency (Debbal and Bereksi-Reguig 2008). Hence, Wavelet Transform, a proven tool for time-frequency analysis, was used. The most important aspect of feature extraction is the selection of a suitable wavelet and the number of levels of decomposition. Usually, the signal is visually inspected first and if they are kind of continuous, Haar or other sharp wavelet functions are used; otherwise a smoother wavelet can be employed. Another method is to perform tests with different types of wavelets and the one which gives maximum efficiency can be selected for the particular application. The number of levels decomposition is chosen based on the dominant frequency components of the signal. The levels are chosen such that those parts of the signal that correlate well with the frequencies required for classification of the signal are retained in the wavelet coefficients. Since the frequency range of heart murmurs is 0 to 1000Hz (Zhou 2008), the number of levels was chosen to be 5. The normalized signal was decomposed into five subbands using Daubechies db6 wavelet decomposition. The DWT decomposition of the input signal x[n] is schematically shown in Figure 4.3. h[n] 2 Level 5 Approximations h[n] 2 h[n] 2 A 2... g[n] 2 A 5 Level 5 Details x[n] A 1 g[n] 2 D 2 Level 2 Details D 5 g[n] 2 Level 1 Details D 1 Figure 4.3 Level-5 wavelet decomposition to obtain wavelet coefficients

8 59 The decomposition of the signal into different frequency bands is simply obtained by successive highpass and lowpass filtering of the signal (Kandaswamy et al 2003; Sadik and Mustafa 2007). The decomposed subbands with their corresponding ranges of frequency are shown in Table 4.2. The resulting wavelet coefficients provide a compact representation of the energy distribution of the signal in each subband. As the frequency spectrum of heart murmurs range from 0 to 1000 Hz and the values of the observed wavelet coefficients in D1 and D2 were also very close to zero, the higher subbands D1 and D2 were discarded from further processing. Table 4.2 Frequency range of sub bands Decomposed sub band Frequency Range(Hz) D D D D D A The time domain representation of the subbands for a normal heart sound file is shown in Figure 4.4. From the wavelet coefficients in each subband (D3, D4, D5 and A5), a set of 15 statistical parameters were derived in order to represent the time-frequency characteristics of the heart murmurs. These are shown in Table 4.3.

9 60 Figure 4.4 Decomposition of normal heart sound with db6 wavelet Table 4.3 Features derived for classification of heart murmurs Sl No.of Description of feature No features 1 Mean of the absolute values of the coefficients in each sub band 4 2 Standard deviation of the wavelet coefficients in each sub band 4 3 Average power of the wavelet coefficients in each sub band 4 4 Ratio of the absolute mean values of adjacent sub bands 3 Total 15 In order to automate the whole feature extraction process, Matlab was linked with Microsoft Excel. The features were then exported to the worksheet directly. These 15 parameters along with the corresponding output (type of murmur) forms the input feature vector for classification (Criley et al 2000; January and Zahra 2007). The dataset used in this work has 441 feature vectors or instances.

10 CLASSIFICATION USING ARTIFICIAL NEURAL NETWORKS Neural Network Classifier Artificial Neural Networks (ANN) are computational models that are patterned after the structure of the human brain. They have the ability to learn mathematical relationships between a series of input (independent) variables and the corresponding output (dependent) variables. This is achieved by training the network with a training dataset consisting of input variables and the known or associated outcomes. Networks are programmed to adjust their internal weights based on the mathematical relationships identified between the inputs and outputs in a dataset. The knowledge gained by the learning experience through training is stored in the form of connection weights, which are used to make decisions on test inputs. Once a network has been trained, it can be used for classification tasks with a separate test data set for validation (Rajendra et al 2003; Jack 1996). H 1 X 1 Y 1 I 1 O 1 Inputs X 2 I 2 H 2 O 8 Y 8 Outputs X 15 I 15 H n Input Layer Hidden Layer Output Layer Figure 4.5 Feedforward neural network to classify heart murmurs

11 62 Among the several existing neural network architectures, the feedforward neural network trained using supervised learning with backpropagation algorithm has been chosen for this work. This model, which is considered the most useful learning model, is being widely used in the biomedical field (William 1995; Maurice and Donna 2006). The feedforward neural network used in this classification work is shown in Figure 4.5. The input layer has 15 nodes representing the 15 input features and the output layer has 8 nodes representing normal and seven different classes of murmurs Encoding of Data for ANN The classification scheme of 1-of-C coding has been used for classifying the heart murmurs into one of the eight output types. For each type of heart murmur, a corresponding output class is associated. The feature vector set, x represents the ANN inputs and the corresponding class once coded, constitutes the ANN outputs. In order to make the neural network training more efficient, the input feature vectors were normalized so that all the values fall in the range between 0 and 1 (Jack 1996; Richard and Michael 2005). The encoding of output is as shown in Table 4.4. Table 4.4 Encoding of output for classification of heart murmurs Sl No. Output vector Classification Normal Early systolic Mid systolic Late systolic Holo systolic Early diastolic Mid diastolic Late diastolic

12 63 The value corresponding to the correct class of output is entered as 1 and other values are entered as zero. The modified input vector is x k, with k = 1, 2,..K where K is the number of heart murmur signals. The output vector associated with x k is denoted by y k Training with Backpropagation Learning As inputs and desired outputs are known, the training is done in a supervised fashion. In the basic back propagation (BP) training algorithm, the weights are moved in the direction of the negative gradient. The BP learning updates the network weights and biases in the direction in which the performance function decreases most rapidly the negative of the gradient. A single iteration of this algorithm can be written as w k+1 = w k α k g k (4.1) where w k is a vector of current weights and biases, g k is the current gradient, and α k is the learning rate. Convergence is sometimes faster if a momentum term is added to the weight update formula. When using momentum, the net is proceeding not in the direction of the gradient, but in the direction of a combination of the current gradient and the previous direction of weight correction. In backpropagation learning with momentum, the weights for training step t+1 are based on the weights at training steps t and t-1. The newff function in Matlab s neural network toolbox was used for the generation of feedforward backpropagation neural network architecture. This newff function has learngdm as the default learning function which chooses initnw initialization function for initializing a layer s weights and biases according to the Nguyen-Widrow initialization algorithm. Learning occurs according to learngdm s learning parameters with their default values as 0.01 for learning rate and 0.9 for momentum constant.

13 Activation Function There are different types of activation functions. The most commonly used are : (i) hyperbolic tangent sigmoid (ii) log sigmoid and (iii) linear. The first criterion of an activation function is that the function must output values in the interval range [0,1]. A second criterion is that the function should output a value close to 1 when sufficiently excited. The sigmoid function meets both the criteria. The performance of these three activation functions were experimented in the neural network models constructed Stopping Criteria Network training continues until a specific terminating condition is satisfied. The terminating condition chosen for the network convergence is the minimum mean squared error (MSE) as defined below : 1 p MSE ( d y ) 2 p m m m 1 (4.2) where p = number of training instances d = desired outputs y = output obtained from the neural network The target for mean squared error set in our work is Training stopped when the value of MSE dropped below EXPERIMENTAL RESULTS AND DISCUSSIONS The training algorithm, activation function and the number of neurons in the hidden layer that provide the best results were determined

14 65 empirically. For this purpose, several feedforward backpropagation neural networks were constructed using various combinations of training algorithms, activation functions and different number of neurons in the hidden layer. The training algorithms used are Levenberg-Marquardt (LM), Gradient Descent with Adaptive learning rate (GDA), Resilient Backpropagation (RP) and Scaled Conjugate Gradient Descent (SCG). The activation functions used are logsigmoid, tansigmoid and purelin. Each network was trained with 331 instances (75% of the dataset with uniform distribution of all classes) and tested with 110 instances (25% of the dataset with uniform distribution of all classes). The performance of these neural network models was measured through accuracy, sensitivity and specificity. The results obtained are presented in Table 4.5. Accuracy, sensitivity and specificity were calculated using the following formulae (Sadik and Mustafa 2007) : TP TN Accuracy (4.3) Total TP Sensitivity TP FN TN Specificity TN FP (4.4) (4.5) where TP = True positive, positive cases classified as positive TN = True negative, negative cases classified as negative FN = False negative, positive cases classified as negative FP = False positive, negative cases classified as positive

15 66 Table 4.5 Performance of different neural network models Model number Training Algorithm Activation Function Number of Epochs Number of neurons in hidden layer Accuracy % Sensitivity % Specificity % 1 LM Logsig LM Logsig LM Logsig LM Logsig LM Purelin LM Purelin LM Purelin LM Purelin LM Tansig LM Tansig LM Tansig LM Tansig GDA Logsig GDA Logsig GDA Logsig GDA Logsig GDA Purelin GDA Purelin GDA Purelin GDA Purelin GDA Tansig GDA Tansig GDA Tansig GDA Tansig RP Logsig RP Logsig RP Logsig RP Logsig RP Purelin RP Purelin RP Purelin RP Purelin RP Tansig RP Tansig RP Tansig RP Tansig SCG Logsig SCG Logsig SCG Logsig SCG Logsig SCG Purelin SCG Purelin SCG Purelin SCG Purelin SCG Tansig SCG Tansig SCG Tansig SCG Tansig

16 Percentage 67 The network models that performed best with each of the four training algorithms were compared. The comparison results are presented in Table 4.6 and illustrated in Figure 4.6. Table 4.6 Performance comparison of training algorithms Training Activation algorithms function No. of neurons in hidden layer Number of Epochs Accuracy % Sensitivity % Specificity % LM Tansig SCG Tansig GDA Tansig RP Tansig LM SCG GDA RP Training Algorithms Accuracy Sensitivity Specificity Figure 4.6 Performance comparison of training algorithms It was observed that the neural network model with LM training algorithm and 26 neurons in the hidden layer showed better performance when compared to the models with other training algorithms. It was also noted that tansigmoid activation function outperforms the other two activation functions. Also, given the speed of convergence of the network during

17 68 training is indicated by the number of epochs, the networks trained with LM algorithm were observed to converge faster than those with SCG, GDA and RP. Hence, the neural network model (Sl. No.10 in Table 4.5) with 26 neurons in the hidden layer, trained with Levenberg-Marquardt (LM) algorithm and tansigmoid activation function for both the hidden and output layer was found to yield the best results for the dataset used. The performance of the proposed neural network classifier is comparable with other classifiers cited in literature and the comparison is presented in Table 4.7. Table 4.7 Performance comparison of different classifiers Reference Classifier No of types of murmurs classified Average Classification Accuracy (%) Faizan et al (2006) MLP-ANN Zhou et al (2008) FBNN Yadi et al (2008) MLP-ANN Proposed MLP-ANN SUMMARY In this chapter, the development of a heart murmur classifier was presented. Discrete Wavelet Transform was used to extract a set of 15 features from the heart murmur signals. Feedforward neural network models with backpropagation learning were developed with different combinations of training algorithms and activation functions to classify the heart murmurs into one of the eight types namely normal, early systolic, mid systolic, late systolic, holo systolic, early diastolic, mid diastolic and late diastolic. The neural network model with Levenberg-Marquardt training algorithm and tansigmoid activation function was found to yield the best performance. The results demonstrate the capability of the developed system as a support tool for physicians in the diagnosis of heart murmurs.

PHONOCARDIOGRAM SIGNAL ANALYSIS FOR MURMUR DIAGNOSING USING SHANNON ENERGY ENVELOP AND SEQUENCED DWT DECOMPOSITION

PHONOCARDIOGRAM SIGNAL ANALYSIS FOR MURMUR DIAGNOSING USING SHANNON ENERGY ENVELOP AND SEQUENCED DWT DECOMPOSITION Journal of Engineering Science and Technology Vol., No. 9 (7) 393-4 School of Engineering, Taylor s University PHONOCARDIOGRAM SIGNAL ANALYSIS FOR MURMUR DIAGNOSING USING SHANNON ENERGY ENVELOP AND SEQUENCED

More information

Question 1 Multiple Choice (8 marks)

Question 1 Multiple Choice (8 marks) Philadelphia University Student Name: Faculty of Engineering Student Number: Dept. of Computer Engineering First Exam, First Semester: 2015/2016 Course Title: Neural Networks and Fuzzy Logic Date: 19/11/2015

More information

COLIC AND MURMURS: AN OVERVIEW

COLIC AND MURMURS: AN OVERVIEW COLIC AND MURMURS: AN OVERVIEW Gunther van Loon, DVM, PhD, Department of Large Animal Internal Medicine, Ghent University, Merelbeke, Belgium Introduction Many horses with colic present with a cardiac

More information

CHAPTER 5 WAVELET BASED DETECTION OF VENTRICULAR ARRHYTHMIAS WITH NEURAL NETWORK CLASSIFIER

CHAPTER 5 WAVELET BASED DETECTION OF VENTRICULAR ARRHYTHMIAS WITH NEURAL NETWORK CLASSIFIER 57 CHAPTER 5 WAVELET BASED DETECTION OF VENTRICULAR ARRHYTHMIAS WITH NEURAL NETWORK CLASSIFIER 5.1 INTRODUCTION The cardiac disorders which are life threatening are the ventricular arrhythmias such as

More information

Neural Network based Heart Arrhythmia Detection and Classification from ECG Signal

Neural Network based Heart Arrhythmia Detection and Classification from ECG Signal Neural Network based Heart Arrhythmia Detection and Classification from ECG Signal 1 M. S. Aware, 2 V. V. Shete *Dept. of Electronics and Telecommunication, *MIT College Of Engineering, Pune Email: 1 mrunal_swapnil@yahoo.com,

More information

Heart Murmur Recognition Based on Hidden Markov Model

Heart Murmur Recognition Based on Hidden Markov Model Journal of Signal and Information Processing, 2013, 4, 140-144 http://dx.doi.org/10.4236/jsip.2013.42020 Published Online May 2013 (http://www.scirp.org/journal/jsip) Heart Murmur Recognition Based on

More information

Automatic Detection of Heart Disease Using Discreet Wavelet Transform and Artificial Neural Network

Automatic Detection of Heart Disease Using Discreet Wavelet Transform and Artificial Neural Network e-issn: 2349-9745 p-issn: 2393-8161 Scientific Journal Impact Factor (SJIF): 1.711 International Journal of Modern Trends in Engineering and Research www.ijmter.com Automatic Detection of Heart Disease

More information

Classıfıcatıon of Dıabetes Dısease Usıng Backpropagatıon and Radıal Basıs Functıon Network

Classıfıcatıon of Dıabetes Dısease Usıng Backpropagatıon and Radıal Basıs Functıon Network UTM Computing Proceedings Innovations in Computing Technology and Applications Volume 2 Year: 2017 ISBN: 978-967-0194-95-0 1 Classıfıcatıon of Dıabetes Dısease Usıng Backpropagatıon and Radıal Basıs Functıon

More information

Multilayer Perceptron Neural Network Classification of Malignant Breast. Mass

Multilayer Perceptron Neural Network Classification of Malignant Breast. Mass Multilayer Perceptron Neural Network Classification of Malignant Breast Mass Joshua Henry 12/15/2017 henry7@wisc.edu Introduction Breast cancer is a very widespread problem; as such, it is likely that

More information

CHAPTER-IV DECISION SUPPORT SYSTEM FOR CONGENITAL HEART SEPTUM DEFECT DIAGNOSIS BASED ON ECG SIGNAL FEATURES USING NEURAL NETWORKS

CHAPTER-IV DECISION SUPPORT SYSTEM FOR CONGENITAL HEART SEPTUM DEFECT DIAGNOSIS BASED ON ECG SIGNAL FEATURES USING NEURAL NETWORKS CHAPTER-IV DECISION SUPPORT SYSTEM FOR CONGENITAL HEART SEPTUM DEFECT DIAGNOSIS BASED ON ECG SIGNAL FEATURES USING NEURAL NETWORKS 4.1 Introduction One of the clinical tests performed to diagnose Congenital

More information

2. The heart sounds are produced by a summed series of mechanical events, as follows:

2. The heart sounds are produced by a summed series of mechanical events, as follows: Heart Sounds. Phonocardiography 1 Objectives 1. Phonocardiography - Definition 2. What produces the heart sounds 3. Where to listen for the heart sounds 4. How to record a phonocardiogram 5. Normal heart

More information

Keywords Artificial Neural Networks (ANN), Echocardiogram, BPNN, RBFNN, Classification, survival Analysis.

Keywords Artificial Neural Networks (ANN), Echocardiogram, BPNN, RBFNN, Classification, survival Analysis. Design of Classifier Using Artificial Neural Network for Patients Survival Analysis J. D. Dhande 1, Dr. S.M. Gulhane 2 Assistant Professor, BDCE, Sevagram 1, Professor, J.D.I.E.T, Yavatmal 2 Abstract The

More information

CARDIAC EXAMINATION MINI-QUIZ

CARDIAC EXAMINATION MINI-QUIZ CARDIAC EXAMINATION MINI-QUIZ 1. Sitting bolt upright, your dyspneic (short of breath) patient has visible jugular venous pulsations to the angle of his jaw, which is 12 cm above his sternal angle. What

More information

REVIEW ON ARRHYTHMIA DETECTION USING SIGNAL PROCESSING

REVIEW ON ARRHYTHMIA DETECTION USING SIGNAL PROCESSING REVIEW ON ARRHYTHMIA DETECTION USING SIGNAL PROCESSING Vishakha S. Naik Dessai Electronics and Telecommunication Engineering Department, Goa College of Engineering, (India) ABSTRACT An electrocardiogram

More information

Isolation of Systolic Heart Murmurs Using Wavelet Transform and Energy Index

Isolation of Systolic Heart Murmurs Using Wavelet Transform and Energy Index 28 Congress on Image and Signal Processing Isolation of Syslic Heart Murmurs Using Wavelet Transform and Energy Index Nikolay Atanasov and Taikang Ning Trinity College, Connecticut, USA nikolay.atanasov@trincoll.edu

More information

PHONOCARDIOGRAPHY (PCG)

PHONOCARDIOGRAPHY (PCG) PHONOCARDIOGRAPHY (PCG) The technique of listening to sounds produced by the organs and vessels of the body is called auscultation. The areas at which the heart sounds are heard better are called auscultation

More information

Heart Sounds Parameter Extraction for Automatic Diagnosis

Heart Sounds Parameter Extraction for Automatic Diagnosis , pp.107-112 http://dx.doi.org/10.14257/astl.205.97.18 Heart Sounds Parameter Extraction for Automatic Diagnosis António Meireles 1, Lino Figueiredo 1, and Luís Seabra Lopes 2 1 GECAD - Knowledge Engineering

More information

Minimum Feature Selection for Epileptic Seizure Classification using Wavelet-based Feature Extraction and a Fuzzy Neural Network

Minimum Feature Selection for Epileptic Seizure Classification using Wavelet-based Feature Extraction and a Fuzzy Neural Network Appl. Math. Inf. Sci. 8, No. 3, 129-1300 (201) 129 Applied Mathematics & Information Sciences An International Journal http://dx.doi.org/10.1278/amis/0803 Minimum Feature Selection for Epileptic Seizure

More information

INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATION ENGINEERING & TECHNOLOGY (IJECET)

INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATION ENGINEERING & TECHNOLOGY (IJECET) INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATION ENGINEERING & TECHNOLOGY (IJECET) International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN 0976 ISSN 0976 6464(Print)

More information

An ECG Beat Classification Using Adaptive Neuro- Fuzzy Inference System

An ECG Beat Classification Using Adaptive Neuro- Fuzzy Inference System An ECG Beat Classification Using Adaptive Neuro- Fuzzy Inference System Pramod R. Bokde Department of Electronics Engineering, Priyadarshini Bhagwati College of Engineering, Nagpur, India Abstract Electrocardiography

More information

NEURAL NETWORK CLASSIFICATION OF EEG SIGNAL FOR THE DETECTION OF SEIZURE

NEURAL NETWORK CLASSIFICATION OF EEG SIGNAL FOR THE DETECTION OF SEIZURE NEURAL NETWORK CLASSIFICATION OF EEG SIGNAL FOR THE DETECTION OF SEIZURE Shaguftha Yasmeen, M.Tech (DEC), Dept. of E&C, RIT, Bangalore, shagufthay@gmail.com Dr. Maya V Karki, Professor, Dept. of E&C, RIT,

More information

Applying Data Mining for Epileptic Seizure Detection

Applying Data Mining for Epileptic Seizure Detection Applying Data Mining for Epileptic Seizure Detection Ying-Fang Lai 1 and Hsiu-Sen Chiang 2* 1 Department of Industrial Education, National Taiwan Normal University 162, Heping East Road Sec 1, Taipei,

More information

Efficient Computation of Phonocardiographic Signal Analysis in Digital Signal Processor Based System

Efficient Computation of Phonocardiographic Signal Analysis in Digital Signal Processor Based System Efficient Computation of Phonocardiographic Signal Analysis in Digital Signal Processor Based System D. Balasubramaniam 1 and D. Nedumaran 2 Abstract This article presents a real-time and cost effective

More information

EEG signal classification using Bayes and Naïve Bayes Classifiers and extracted features of Continuous Wavelet Transform

EEG signal classification using Bayes and Naïve Bayes Classifiers and extracted features of Continuous Wavelet Transform EEG signal classification using Bayes and Naïve Bayes Classifiers and extracted features of Continuous Wavelet Transform Reza Yaghoobi Karimoi*, Mohammad Ali Khalilzadeh, Ali Akbar Hossinezadeh, Azra Yaghoobi

More information

Heart sounds and murmurs. Dr. Szathmári Miklós Semmelweis University First Department of Medicine 15. Oct

Heart sounds and murmurs. Dr. Szathmári Miklós Semmelweis University First Department of Medicine 15. Oct Heart sounds and murmurs Dr. Szathmári Miklós Semmelweis University First Department of Medicine 15. Oct. 2013. Conditions for auscultation of the heart Quiet room Patient comfortable Chest fully exposed

More information

Brain Tumour Diagnostic Support Based on Medical Image Segmentation

Brain Tumour Diagnostic Support Based on Medical Image Segmentation Brain Tumour Diagnostic Support Based on Medical Image Segmentation Z. Měřínský, E. Hošťálková, A. Procházka Institute of Chemical Technology, Prague Department of Computing and Control Engineering Abstract

More information

Extraction of Unwanted Noise in Electrocardiogram (ECG) Signals Using Discrete Wavelet Transformation

Extraction of Unwanted Noise in Electrocardiogram (ECG) Signals Using Discrete Wavelet Transformation Extraction of Unwanted Noise in Electrocardiogram (ECG) Signals Using Discrete Wavelet Transformation Er. Manpreet Kaur 1, Er. Gagandeep Kaur 2 M.Tech (CSE), RIMT Institute of Engineering & Technology,

More information

Application of Artificial Neural Networks in Classification of Autism Diagnosis Based on Gene Expression Signatures

Application of Artificial Neural Networks in Classification of Autism Diagnosis Based on Gene Expression Signatures Application of Artificial Neural Networks in Classification of Autism Diagnosis Based on Gene Expression Signatures 1 2 3 4 5 Kathleen T Quach Department of Neuroscience University of California, San Diego

More information

Quick detection of QRS complexes and R-waves using a wavelet transform and K-means clustering

Quick detection of QRS complexes and R-waves using a wavelet transform and K-means clustering Bio-Medical Materials and Engineering 26 (2015) S1059 S1065 DOI 10.3233/BME-151402 IOS Press S1059 Quick detection of QRS complexes and R-waves using a wavelet transform and K-means clustering Yong Xia

More information

Classification of heart signal using wavelet haar and backpropagation neural network

Classification of heart signal using wavelet haar and backpropagation neural network IOP Conference Series: Materials Science and Engineering PAPER OPEN ACCESS Classification of heart signal using wavelet haar and backpropagation neural network To cite this article: H Hindarto et al 28

More information

EEG Signal Classification Using Wavelet Feature Extraction and Neural Networks

EEG Signal Classification Using Wavelet Feature Extraction and Neural Networks EEG Signal Classification Using Wavelet Feature Extraction and Neural Networks Pari Jahankhani, Vassilis Kodogiannis and Kenneth Revett AbstractDecision Support Systems have been utilised since 196, providing

More information

Wavelet Decomposition for Detection and Classification of Critical ECG Arrhythmias

Wavelet Decomposition for Detection and Classification of Critical ECG Arrhythmias Proceedings of the 8th WSEAS Int. Conference on Mathematics and Computers in Biology and Chemistry, Vancouver, Canada, June 19-21, 2007 80 Wavelet Decomposition for Detection and Classification of Critical

More information

HISTORY. Question: What category of heart disease is suggested by this history? CHIEF COMPLAINT: Heart murmur present since early infancy.

HISTORY. Question: What category of heart disease is suggested by this history? CHIEF COMPLAINT: Heart murmur present since early infancy. HISTORY 18-year-old man. CHIEF COMPLAINT: Heart murmur present since early infancy. PRESENT ILLNESS: Although normal at birth, a heart murmur was heard at the six week check-up and has persisted since

More information

PC based Heart Sound Monitoring System

PC based Heart Sound Monitoring System PC based Heart Sound Monitoring System Arathy R Assistant Professor in Electronics and biomedical Engineering Model Engineering College Thrikkakara, Cochin Gowriprabha V PGET-D&D,Skanray Healthcare pvt

More information

Monitoring Cardiac Stress Using Features Extracted From S1 Heart Sounds

Monitoring Cardiac Stress Using Features Extracted From S1 Heart Sounds e-issn 2455 1392 Volume 2 Issue 4, April 2016 pp. 271-275 Scientific Journal Impact Factor : 3.468 http://www.ijcter.com Monitoring Cardiac Stress Using Features Extracted From S1 Heart Sounds Biju V.

More information

A Novel Application of Wavelets to Real-Time Detection of R-waves

A Novel Application of Wavelets to Real-Time Detection of R-waves A Novel Application of Wavelets to Real-Time Detection of R-waves Katherine M. Davis,* Richard Ulrich and Antonio Sastre I Introduction In recent years, medical, industrial and military institutions have

More information

Classification of Cardiac Arrhythmias based on Dual Tree Complex Wavelet Transform

Classification of Cardiac Arrhythmias based on Dual Tree Complex Wavelet Transform Classification of Cardiac Arrhythmias based on Dual Tree Complex Wavelet Transform Manu Thomas, Manab Kr Das Student Member, IEEE and Samit Ari, Member, IEEE Abstract The electrocardiogram (ECG) is a standard

More information

Heart Diseases diagnosis using intelligent algorithm based on PCG signal analysis

Heart Diseases diagnosis using intelligent algorithm based on PCG signal analysis Heart Diseases diagnosis using intelligent algorithm based on PCG signal analysis Mohammed Nabih-Ali 1, EL-Sayed A. El-Dahshan 1, 2 and Ashraf S. Yahia 2 1 Egyptian E-Learning University (EELU), 33 El-meshah

More information

Physiology of the Heart Delmar Learning, a Division of Thomson Learning, Inc.

Physiology of the Heart Delmar Learning, a Division of Thomson Learning, Inc. Physiology of the Heart 2004 Delmar Learning, a Division of Thomson Learning, Inc. Physiology of the Heart State Standards 35) Outline the structure and functions of the anatomy of the cardiovascular system,

More information

DISCRETE WAVELET PACKET TRANSFORM FOR ELECTROENCEPHALOGRAM- BASED EMOTION RECOGNITION IN THE VALENCE-AROUSAL SPACE

DISCRETE WAVELET PACKET TRANSFORM FOR ELECTROENCEPHALOGRAM- BASED EMOTION RECOGNITION IN THE VALENCE-AROUSAL SPACE DISCRETE WAVELET PACKET TRANSFORM FOR ELECTROENCEPHALOGRAM- BASED EMOTION RECOGNITION IN THE VALENCE-AROUSAL SPACE Farzana Kabir Ahmad*and Oyenuga Wasiu Olakunle Computational Intelligence Research Cluster,

More information

ISPUB.COM. Spectral analysis of the PCG signals. S Debbal, F Bereksi-Reguig INTRODUCTION THEORITICAL BACKGROUND RESULTS AND DISCUSSION.

ISPUB.COM. Spectral analysis of the PCG signals. S Debbal, F Bereksi-Reguig INTRODUCTION THEORITICAL BACKGROUND RESULTS AND DISCUSSION. ISPUB.COM The Internet Journal of Medical Technology Volume 4 Number 1 Spectral analysis of the PCG signals S Debbal, F Bereksi-Reguig Citation S Debbal, F Bereksi-Reguig. Spectral analysis of the PCG

More information

Auto-Encoder Pre-Training of Segmented-Memory Recurrent Neural Networks

Auto-Encoder Pre-Training of Segmented-Memory Recurrent Neural Networks Auto-Encoder Pre-Training of Segmented-Memory Recurrent Neural Networks Stefan Glüge, Ronald Böck and Andreas Wendemuth Faculty of Electrical Engineering and Information Technology Cognitive Systems Group,

More information

Automatic Detection of Epileptic Seizures in EEG Using Machine Learning Methods

Automatic Detection of Epileptic Seizures in EEG Using Machine Learning Methods Automatic Detection of Epileptic Seizures in EEG Using Machine Learning Methods Ying-Fang Lai 1 and Hsiu-Sen Chiang 2* 1 Department of Industrial Education, National Taiwan Normal University 162, Heping

More information

Priya Rani 1, A N Cheeran 2, Vaibhav D Awandekar 3 and Rameshwari S Mane 4

Priya Rani 1, A N Cheeran 2, Vaibhav D Awandekar 3 and Rameshwari S Mane 4 Remote Monitoring of Heart Sounds in Real-Time Priya Rani 1, A N Cheeran 2, Vaibhav D Awandekar 3 and Rameshwari S Mane 4 1,4 M. Tech. Student (Electronics), VJTI, Mumbai, Maharashtra 2 Associate Professor,

More information

Classification of Real Heart Disease Using Probabilistic Neural Network

Classification of Real Heart Disease Using Probabilistic Neural Network Impact Factor (SJIF): 5.301 International Journal of Advance Research in Engineering, Science & Technology e-issn: 2393-9877, p-issn: 2394-2444 Volume 5, Issue 9, September-2018 Classification of Real

More information

Removal of Baseline wander and detection of QRS complex using wavelets

Removal of Baseline wander and detection of QRS complex using wavelets International Journal of Scientific & Engineering Research Volume 3, Issue 4, April-212 1 Removal of Baseline wander and detection of QRS complex using wavelets Nilesh Parihar, Dr. V. S. Chouhan Abstract

More information

COMPARATIVE STUDY ON FEATURE EXTRACTION METHOD FOR BREAST CANCER CLASSIFICATION

COMPARATIVE STUDY ON FEATURE EXTRACTION METHOD FOR BREAST CANCER CLASSIFICATION COMPARATIVE STUDY ON FEATURE EXTRACTION METHOD FOR BREAST CANCER CLASSIFICATION 1 R.NITHYA, 2 B.SANTHI 1 Asstt Prof., School of Computing, SASTRA University, Thanjavur, Tamilnadu, India-613402 2 Prof.,

More information

Genetic Algorithm based Feature Extraction for ECG Signal Classification using Neural Network

Genetic Algorithm based Feature Extraction for ECG Signal Classification using Neural Network Genetic Algorithm based Feature Extraction for ECG Signal Classification using Neural Network 1 R. Sathya, 2 K. Akilandeswari 1,2 Research Scholar 1 Department of Computer Science 1 Govt. Arts College,

More information

Classification of benign and malignant masses in breast mammograms

Classification of benign and malignant masses in breast mammograms Classification of benign and malignant masses in breast mammograms A. Šerifović-Trbalić*, A. Trbalić**, D. Demirović*, N. Prljača* and P.C. Cattin*** * Faculty of Electrical Engineering, University of

More information

FIR filter bank design for Audiogram Matching

FIR filter bank design for Audiogram Matching FIR filter bank design for Audiogram Matching Shobhit Kumar Nema, Mr. Amit Pathak,Professor M.Tech, Digital communication,srist,jabalpur,india, shobhit.nema@gmail.com Dept.of Electronics & communication,srist,jabalpur,india,

More information

A Novel Design and Development of Condenser Microphone Based Stethoscope to Analyze Phonocardiogram Spectrum Using Audacity

A Novel Design and Development of Condenser Microphone Based Stethoscope to Analyze Phonocardiogram Spectrum Using Audacity A Novel Design and Development of Condenser Microphone Based Stethoscope to Analyze Phonocardiogram Spectrum Using Audacity Tarak Das 1, Parameswari Hore 1, Prarthita Sharma 1, Tapas Kr. Dawn 2 1 Department

More information

Electrocardiogram beat classification using Discrete Wavelet Transform, higher order statistics and multivariate analysis

Electrocardiogram beat classification using Discrete Wavelet Transform, higher order statistics and multivariate analysis Electrocardiogram beat classification using Discrete Wavelet Transform, higher order statistics and multivariate analysis Thripurna Thatipelli 1, Padmavathi Kora 2 1Assistant Professor, Department of ECE,

More information

The Cardiovascular System Part I: Heart Outline of class lecture After studying part I of this chapter you should be able to:

The Cardiovascular System Part I: Heart Outline of class lecture After studying part I of this chapter you should be able to: The Cardiovascular System Part I: Heart Outline of class lecture After studying part I of this chapter you should be able to: 1. Describe the functions of the heart 2. Describe the location of the heart,

More information

Detection of pulmonary abnormalities using Multi scale products and ARMA modelling

Detection of pulmonary abnormalities using Multi scale products and ARMA modelling Volume 119 No. 15 2018, 2177-2181 ISSN: 1314-3395 (on-line version) url: http://www.acadpubl.eu/hub/ http://www.acadpubl.eu/hub/ Detection of pulmonary abnormalities using Multi scale products and ARMA

More information

COMPRESSED ECG BIOMETRIC USING CARDIOID GRAPH BASED FEATURE EXTRACTION

COMPRESSED ECG BIOMETRIC USING CARDIOID GRAPH BASED FEATURE EXTRACTION COMPRESSED ECG BIOMETRIC USING CARDIOID GRAPH BASED FEATURE EXTRACTION Fatema-tuz-Zohra and Khairul Azami Sidek Department of Electrical of Electrical and Computer Engineering Faculty of Engineering, International

More information

CARDIAC ARRYTHMIA CLASSIFICATION BY NEURONAL NETWORKS (MLP)

CARDIAC ARRYTHMIA CLASSIFICATION BY NEURONAL NETWORKS (MLP) CARDIAC ARRYTHMIA CLASSIFICATION BY NEURONAL NETWORKS (MLP) Bochra TRIQUI, Abdelkader BENYETTOU Center for Artificial Intelligent USTO-MB University Algeria triqui_bouchra@yahoo.fr a_benyettou@yahoo.fr

More information

EEG ANALYSIS: ANN APPROACH

EEG ANALYSIS: ANN APPROACH EEG ANALYSIS: ANN APPROACH CHAPTER 5 EEG ANALYSIS: ANN APPROACH 5.1 INTRODUCTION The analysis of EEG signals using ANN deals with developing a network in order to establish a relation between input and

More information

Coimbatore , India. 2 Professor, Department of Information Technology, PSG College of Technology, Coimbatore , India.

Coimbatore , India. 2 Professor, Department of Information Technology, PSG College of Technology, Coimbatore , India. Research Paper OPTIMAL SELECTION OF FEATURE EXTRACTION METHOD FOR PNN BASED AUTOMATIC CARDIAC ARRHYTHMIA CLASSIFICATION Rekha.R 1,* and Vidhyapriya.R 2 Address for Correspondence 1 Assistant Professor,

More information

Fuzzy Based Early Detection of Myocardial Ischemia Using Wavelets

Fuzzy Based Early Detection of Myocardial Ischemia Using Wavelets Fuzzy Based Early Detection of Myocardial Ischemia Using Wavelets Jyoti Arya 1, Bhumika Gupta 2 P.G. Student, Department of Computer Science, GB Pant Engineering College, Ghurdauri, Pauri, India 1 Assistant

More information

Frequency Tracking: LMS and RLS Applied to Speech Formant Estimation

Frequency Tracking: LMS and RLS Applied to Speech Formant Estimation Aldebaro Klautau - http://speech.ucsd.edu/aldebaro - 2/3/. Page. Frequency Tracking: LMS and RLS Applied to Speech Formant Estimation ) Introduction Several speech processing algorithms assume the signal

More information

Cardiac Cycle MCQ. Professor of Cardiovascular Physiology. Cairo University 2007

Cardiac Cycle MCQ. Professor of Cardiovascular Physiology. Cairo University 2007 Cardiac Cycle MCQ Abdel Moniem Ibrahim Ahmed, MD Professor of Cardiovascular Physiology Cairo University 2007 1- Regarding the length of systole and diastole: a- At heart rate 75 b/min, the duration of

More information

Murmur Sounds made by turbulence in the heart or blood stream. 1. Timing. 5. Intensity 2. Shape. 6. Pitch 3. Location of maximum intensity

Murmur Sounds made by turbulence in the heart or blood stream. 1. Timing. 5. Intensity 2. Shape. 6. Pitch 3. Location of maximum intensity Definition Items in description of Timing Shape Location of maximum intensity Murmur Sounds made by turbulence in the heart or blood stream. 1. Timing 5. Intensity 2. Shape 6. Pitch 3. Location of maximum

More information

HEART MURMURS: DECIPHERING THEIR CAUSE AND SIGNIFICANCE

HEART MURMURS: DECIPHERING THEIR CAUSE AND SIGNIFICANCE Vet Times The website for the veterinary profession https://www.vettimes.co.uk HEART MURMURS: DECIPHERING THEIR CAUSE AND SIGNIFICANCE Author : Pedro Oliveira Categories : Vets Date : May 27, 2013 PEDRO

More information

Clinical significance of cardiac murmurs: Get the sound and rhythm!

Clinical significance of cardiac murmurs: Get the sound and rhythm! Clinical significance of cardiac murmurs: Get the sound and rhythm! Prof. dr. Gunther van Loon, DVM, PhD, Ass Member ECVDI, Dip ECEIM Dept. of Large Animal Internal Medicine Ghent University, Belgium Murmurs

More information

CARDIOVASCULAR PHYSIOLOGY

CARDIOVASCULAR PHYSIOLOGY CARDIOVASCULAR PHYSIOLOGY LECTURE 4 Cardiac cycle Polygram - analysis of cardiac activity Ana-Maria Zagrean MD, PhD The Cardiac Cycle - definitions: the sequence of electrical and mechanical events that

More information

CHAPTER IV PREPROCESSING & FEATURE EXTRACTION IN ECG SIGNALS

CHAPTER IV PREPROCESSING & FEATURE EXTRACTION IN ECG SIGNALS CHAPTER IV PREPROCESSING & FEATURE EXTRACTION IN ECG SIGNALS are The proposed ECG classification approach consists of three phases. They Preprocessing Feature Extraction and Selection Classification The

More information

PART I: HEART ANATOMY

PART I: HEART ANATOMY Lab 7: Heart Sounds and Blood Pressure PART I: HEART ANATOMY a) You should be able to identify the following structures on an adult human heart diagram. the 4 chambers the bicuspid (mitral) and tricuspid

More information

Classifying heart sounds using peak location for segmentation and feature construction

Classifying heart sounds using peak location for segmentation and feature construction Classifying heart sounds using peak location for segmentation and feature construction Elsa Ferreira Gomes GECAD - Knowledge Eng.Decision Support Institute of Engineering (ISEP/IPP) Porto, Portugal Emanuel

More information

Objectives of the Heart

Objectives of the Heart Objectives of the Heart Electrical activity of the heart Action potential EKG Cardiac cycle Heart sounds Heart Rate The heart s beat separated into 2 phases Relaxed phase diastole (filling of the chambers)

More information

Discrete Wavelet Transform-based Baseline Wandering Removal for High Resolution Electrocardiogram

Discrete Wavelet Transform-based Baseline Wandering Removal for High Resolution Electrocardiogram 26 C. Bunluechokchai and T. Leeudomwong: Discrete Wavelet Transform-based Baseline... (26-31) Discrete Wavelet Transform-based Baseline Wandering Removal for High Resolution Electrocardiogram Chissanuthat

More information

Study and Design of a Shannon-Energy-Envelope based Phonocardiogram Peak Spacing Analysis for Estimating Arrhythmic Heart-Beat

Study and Design of a Shannon-Energy-Envelope based Phonocardiogram Peak Spacing Analysis for Estimating Arrhythmic Heart-Beat International Journal of Scientific and Research Publications, Volume 4, Issue 9, September 2014 1 Study and Design of a Shannon-Energy-Envelope based Phonocardiogram Peak Spacing Analysis for Estimating

More information

Research Article Artificial Neural Network-Based Automated ECG Signal Classifier

Research Article Artificial Neural Network-Based Automated ECG Signal Classifier ISRN Biomedical Engineering Volume 23, Article ID 2697, 6 pages http://dx.doi.org/5/23/2697 Research Article Artificial Neural Network-Based Automated ECG Signal Classifier Sahar H. El-Khafif,2 and Mohamed

More information

An Edge-Device for Accurate Seizure Detection in the IoT

An Edge-Device for Accurate Seizure Detection in the IoT An Edge-Device for Accurate Seizure Detection in the IoT M. A. Sayeed 1, S. P. Mohanty 2, E. Kougianos 3, and H. Zaveri 4 University of North Texas, Denton, TX, USA. 1,2,3 Yale University, New Haven, CT,

More information

One-Year Survival Prediction of Myocardial Infarction

One-Year Survival Prediction of Myocardial Infarction One-Year Survival Prediction of Myocardial Infarction 1 Abdulkader Helwan, 2 Dilber Uzun Ozsahin 1,2 Department of Biomedical Engineering, Near East University, Near East Boulevard, TRNC, Nicosia, 99138

More information

A framework for automatic heart sound analysis without segmentation

A framework for automatic heart sound analysis without segmentation RESEARCH Open Access A framework for automatic heart sound analysis without segmentation Sumeth Yuenyong 1*, Akinori Nishihara 1, Waree Kongprawechnon 2, Kanokvate Tungpimolrut 3 * Correspondence: toey123@gmail.

More information

Accurate Prediction of Heart Disease Diagnosing Using Computation Method

Accurate Prediction of Heart Disease Diagnosing Using Computation Method Accurate Prediction of Heart Disease Diagnosing Using Computation Method 1 Hanumanthappa H, 2 Pundalik Chavan 1 Assistant Professor, 2 Assistant Professor 1 Computer Science & Engineering, 2 Computer Science

More information

ECG Rhythm Analysis by Using Neuro-Genetic Algorithms

ECG Rhythm Analysis by Using Neuro-Genetic Algorithms MASAUM Journal of Basic and Applied Sciences, Vol. 1, No. 3, October 2009 522 ECG Rhythm Analysis by Using Neuro-Genetic Algorithms Safaa S. Omran, S.M.R. Taha, and Nassr Ali Awadh Abstract The heart is

More information

ECG Signal Analysis for Abnormality Detection in the Heart beat

ECG Signal Analysis for Abnormality Detection in the Heart beat GRD Journals- Global Research and Development Journal for Engineering Volume 1 Issue 10 September 2016 ISSN: 2455-5703 ECG Signal Analysis for Abnormality Detection in the Heart beat Vedprakash Gujiri

More information

ANN predicts locoregional control using molecular marker profiles of. Head and Neck squamous cell carcinoma

ANN predicts locoregional control using molecular marker profiles of. Head and Neck squamous cell carcinoma ANN predicts locoregional control using molecular marker profiles of Head and Neck squamous cell carcinoma Final Project: 539 Dinesh Kumar Tewatia Introduction Radiotherapy alone or combined with chemotherapy,

More information

1. how a careful cardiovascular evaluation can accurately assess pathology and physiology at the bedside, and

1. how a careful cardiovascular evaluation can accurately assess pathology and physiology at the bedside, and This program will demonstrate: 1. how a careful cardiovascular evaluation can accurately assess pathology and physiology at the bedside, and 2. the importance of integrating this information with selected

More information

A HMM-based Pre-training Approach for Sequential Data

A HMM-based Pre-training Approach for Sequential Data A HMM-based Pre-training Approach for Sequential Data Luca Pasa 1, Alberto Testolin 2, Alessandro Sperduti 1 1- Department of Mathematics 2- Department of Developmental Psychology and Socialisation University

More information

Cardiac Ausculation in the Elderly

Cardiac Ausculation in the Elderly Cardiac Ausculation in the Elderly 박성하 신촌세브란스병원심장혈관병원심장내과 Anatomy Surface projection of the Heart and Great Vessels Evaluating pulsation Superior vena cava Rt. pulmonary artery Right atrium Right ventricle

More information

A Novel Fourth Heart Sounds Segmentation Algorithm Based on Matching Pursuit and Gabor Dictionaries

A Novel Fourth Heart Sounds Segmentation Algorithm Based on Matching Pursuit and Gabor Dictionaries A Novel Fourth Heart Sounds Segmentation Algorithm Based on Matching Pursuit and Gabor Dictionaries Carlos Nieblas 1, Roilhi Ibarra 2 and Miguel Alonso 2 1 Meltsan Solutions, Blvd. Adolfo Lopez Mateos

More information

Prediction of Diabetes by using Artificial Neural Network

Prediction of Diabetes by using Artificial Neural Network 2011 International Conference on Circuits, System and Simulation IPCSIT vol.7 (2011) (2011) IACSIT Press, Singapore Prediction of Diabetes by using Artificial Neural Network Muhammad Akmal Sapon 1, Khadijah

More information

Performance Analysis of Epileptic EEG Expert System Using Scaled Conjugate Back Propagation Based ANN Classifier

Performance Analysis of Epileptic EEG Expert System Using Scaled Conjugate Back Propagation Based ANN Classifier Performance Analysis of Epileptic EEG Expert System Using Scaled Conjugate Back Propagation Based ANN Classifier Ashish Raj, Pankaj Gakhar, Meenu Kumari 3 Sweta Tripathi 4 ashishraj987@gmail.com,pankaj.gakhar@poornima.edu.in,meenu.kumari@poornima.edu.in

More information

CSE Introduction to High-Perfomance Deep Learning ImageNet & VGG. Jihyung Kil

CSE Introduction to High-Perfomance Deep Learning ImageNet & VGG. Jihyung Kil CSE 5194.01 - Introduction to High-Perfomance Deep Learning ImageNet & VGG Jihyung Kil ImageNet Classification with Deep Convolutional Neural Networks Alex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton,

More information

Artificial neural networks: application to electrical stimulation of the human nervous system

Artificial neural networks: application to electrical stimulation of the human nervous system Artificial neural networks: application to electrical stimulation of the human nervous system Richard B. North, M.D., J. Paul McNamee, M.S., Lee Wu, M.S., and Steven Piantadosi, M.D., Ph.D. Departments

More information

Detection and Classification of QRS and ST segment using WNN

Detection and Classification of QRS and ST segment using WNN Detection and Classification of QRS and ST segment using WNN 1 Surendra Dalu, 2 Nilesh Pawar 1 Electronics and Telecommunication Department, Government polytechnic Amravati, Maharastra, 44461, India 2

More information

Do Now. Get out work from last class to be checked

Do Now. Get out work from last class to be checked Do Now Get out work from last class to be checked Heart Actions Cardiac Cycle: One complete heartbeat. The contraction of a heart chamber is called systole and the relaxation of a chamber is called diastole.

More information

Lab #3: Electrocardiogram (ECG / EKG)

Lab #3: Electrocardiogram (ECG / EKG) Lab #3: Electrocardiogram (ECG / EKG) An introduction to the recording and analysis of cardiac activity Introduction The beating of the heart is triggered by an electrical signal from the pacemaker. The

More information

A Learning Method of Directly Optimizing Classifier Performance at Local Operating Range

A Learning Method of Directly Optimizing Classifier Performance at Local Operating Range A Learning Method of Directly Optimizing Classifier Performance at Local Operating Range Lae-Jeong Park and Jung-Ho Moon Department of Electrical Engineering, Kangnung National University Kangnung, Gangwon-Do,

More information

ANALYSIS AND CLASSIFICATION OF HEART SOUNDS AND MURMETRS BASED ON TEE INSTANTANEOUS ENERGY AND FREQUENCY ESTIMATIONS

ANALYSIS AND CLASSIFICATION OF HEART SOUNDS AND MURMETRS BASED ON TEE INSTANTANEOUS ENERGY AND FREQUENCY ESTIMATIONS ANALYSIS AND CLASSIFICATION OF HEART SOUNDS AND MURMETRS BASED ON TEE INSTANTANEOUS ENERGY AND FREQUENCY ESTIMATIONS Zaiton Sharg MOM S him &id, Ahmad Zuri Sha ameri, Sheikh H in Shaikh sblleh Digital

More information

Accuracy Enhancement of the Epileptic Seizure Detection in EEG Signals

Accuracy Enhancement of the Epileptic Seizure Detection in EEG Signals Appl. Math. Inf. Sci. 11, No. 6, 1693-172 (217) 1693 Applied Mathematics & Information Sciences An International Journal http://dx.doi.org/1.18576/amis/11616 Accuracy Enhancement of the Epileptic Seizure

More information

Development of 2-Channel Eeg Device And Analysis Of Brain Wave For Depressed Persons

Development of 2-Channel Eeg Device And Analysis Of Brain Wave For Depressed Persons Development of 2-Channel Eeg Device And Analysis Of Brain Wave For Depressed Persons P.Amsaleka*, Dr.S.Mythili ** * PG Scholar, Applied Electronics, Department of Electronics and Communication, PSNA College

More information

Ischemia classification via ECG using MLP neural networks

Ischemia classification via ECG using MLP neural networks International Journal of Computational Intelligence Systems, Vol. 7, No. 2 (April 2014), 344-352 Ischemia classification via ECG using MLP neural networks Peláez J.I. Department of Languages and Computer

More information

Case # 1. Page: 8. DUKE: Adams

Case # 1. Page: 8. DUKE: Adams Case # 1 Page: 8 1. The cardiac output in this patient is reduced because of: O a) tamponade physiology O b) restrictive physiology O c) coronary artery disease O d) left bundle branch block Page: 8 1.

More information

Noise Cancellation using Adaptive Filters Algorithms

Noise Cancellation using Adaptive Filters Algorithms Noise Cancellation using Adaptive Filters Algorithms Suman, Poonam Beniwal Department of ECE, OITM, Hisar, bhariasuman13@gmail.com Abstract Active Noise Control (ANC) involves an electro acoustic or electromechanical

More information

Sound Quality Evaluation of Hermetic Compressors Using Artificial Neural Networks

Sound Quality Evaluation of Hermetic Compressors Using Artificial Neural Networks Purdue University Purdue e-pubs International Compressor Engineering Conference School of Mechanical Engineering 2006 Sound Quality Evaluation of Hermetic Compressors Using Artificial Neural Networks Claudio

More information

Predicting Breast Cancer Survivability Rates

Predicting Breast Cancer Survivability Rates Predicting Breast Cancer Survivability Rates For data collected from Saudi Arabia Registries Ghofran Othoum 1 and Wadee Al-Halabi 2 1 Computer Science, Effat University, Jeddah, Saudi Arabia 2 Computer

More information

THE CARDIOVASCULAR SYSTEM. Heart 2

THE CARDIOVASCULAR SYSTEM. Heart 2 THE CARDIOVASCULAR SYSTEM Heart 2 PROPERTIES OF CARDIAC MUSCLE Cardiac muscle Striated Short Wide Branched Interconnected Skeletal muscle Striated Long Narrow Cylindrical PROPERTIES OF CARDIAC MUSCLE Intercalated

More information