ECG beat classification by a novel hybrid neural network

Dokur, Zümray; Ölmez, Tamer

doi:10.1016/s0169-2607(00)00133-4

Cited by 181 publications

(78 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The structure is not automatically determined by the training algorithm, 3) Back-propagation algorithm may be caught by local minima, which decreases network performance. We observed these disadvantages in the previous study [8]. Especially, the number of nodes in the hidden layers could be defined after many trials, which took too long time.…”

Section: Artificial Neural Networkmentioning

confidence: 93%

Classification of Respiratory Sounds by Using an Artificial Neural Network

Dokur

Ölmez

2003

Int. J. Patt. Recogn. Artif. Intell.

Self Cite

View full text Add to dashboard Cite

Abstract-In this paper, a classification method for respiratory sounds (RSs) in patients with asthma and in healthy subjects is presented.Wavelet transform is applied to a window containing 256 samples. Elements of the feature vectors are obtained from the wavelet coefficients. The best feature elements are selected by using dynamic programming. Grow and Learn (GAL) neural network is used for the classification.It is observed that RSs of patients (with asthma) and healthy subjects are successfully classified by the GAL network.

show abstract

Section: Artificial Neural Networkmentioning

confidence: 93%

Classification of Respiratory Sounds by Using an Artificial Neural Network

Dokur

Ölmez

2003

Int. J. Patt. Recogn. Artif. Intell.

Self Cite

View full text Add to dashboard Cite

show abstract

“…In the literature, it is observed that new neural network structures have been investigated to increase the classification performances [6,15,16]. In the study, three ANNs are comparatively investigated for the segmentation of MR and CT (Computer Topography) head images.…”

Section: Artificial Neural Networkmentioning

confidence: 99%

Segmentation of MR and CT Images by Using a Quantiser Neural Network

Dokur

Ölmez

2003

Neural Computing & Applications

View full text Add to dashboard Cite

A Quantiser Neural Network (QNN) is proposed for the segmentation of MR and CT images. Elements of a feature vector are formed by image intensities at one neighbourhood of the pixel of interest. QNN is a novel neural network structure, which is trained by genetic algorithms. Each node in the first layer of the QNN forms a hyperplane (HP) in the input space. There is a constraint on the HPs in a QNN.The HP is represented by only one parameter in ddimensional input space. Genetic algorithms are used to find the optimum values of the parameters which represent these nodes. The novel neural network is comparatively examined with a multilayer perceptron and a Kohonen network for the segmentation of MR and CT head images. It is observed that the QNN gives the best classification performance with fewer nodes after a short training time.

show abstract

“…The classification of detected arrhythmias is also a research field of great interest. Several classification approaches have been presented in the literature viz wavelet analysis combined with radial basis function neural networks [10] and non-linear dynamic modeling [11]. In this paper the dataset has been collected from MIT-BIH.…”

Section: Introductionmentioning

confidence: 99%

Analysis and Classification of Cardiac Arrhythmia Using ECG Signals

Bhardwaj¹,

Choudhary²,

Dayama³

2012

IJCA

View full text Add to dashboard Cite

ECG is a graphical record of the electrical tension of heart and has established as one the most important bio-signal used by cardiologists for diagnostic purposes and further to adopt an appropriate course of treatment. The difficulties faced in interpretation of ECG signals forced researchers to study about automatic detection of cardiac arrhythmia disorders. The data analysis techniques using specific computer software could easily interpret complex ECG signals, predict presence or absence of cardiac arrhythmia. This provides real time analysis and further facilitates for timely diagnosis. In this paper, Support Vector Machine (SVM) technique, using LibSVM3.1 has been applied to ECG dataset for arrhythmia classification in five categories. Out of these five categories, one is normal and four are arrhythmic beat categories. The dataset used in this study is 3003 arrhythmic beats out of which 2101 beats (70%) are used for training and remaining 902 beats (30%) have been used for testing purpose. Total performance accuracy is found to be around 95.21 % in this case.

show abstract

ECG beat classification by a novel hybrid neural network

Cited by 181 publications

References 15 publications

Classification of Respiratory Sounds by Using an Artificial Neural Network

Classification of Respiratory Sounds by Using an Artificial Neural Network

Segmentation of MR and CT Images by Using a Quantiser Neural Network

Analysis and Classification of Cardiac Arrhythmia Using ECG Signals

Contact Info

Product

Resources

About