Berat Doğan

ECG beat classification using particle swarm optimization and radial basis function neural network

This paper presents a method for electrocardiogram (ECG) beat classification based on particle swarm optimization (PSO) and radial basis function neural network (RBFNN). Six types of beats including Normal Beat, Premature Ventricular Contraction (PVC), Fusion of Ventricular and Normal Beat (F), Atrial Premature Beat (A), Right Bundle Branch Block Beat (R) and Fusion of Paced and Normal Beat (f) are obtained from the MIT-BIH arrhythmia database. Four morphological features are extracted from each beat after the preprocessing of the selected records. For classification stage of the extracted features, a RBFNN structure which is evolved by particle swarm optimization is used. Several experiments are performed over the test set and it is observed that the proposed method classifies ECG beats with a smaller size of network without making any concessions on the classification performance.

A new metaheuristic for numerical function optimization: Vortex Search algorithm

Abstract In this study, a new single-solution based metaheuristic, namely the Vortex Search (VS) algorithm, is proposed to perform numerical function optimization. The proposed VS algorithm is inspired from the vortex pattern created by the vortical flow of the stirred fluids. To provide a good balance between the explorative and exploitative behavior of a search, the proposed method models its search behavior as a vortex pattern by using an adaptive step size adjustment scheme. The proposed VS algorithm is tested over 50 benchmark mathematical functions and the results are compared to both the single-solution based (Simulated Annealing, SA and Pattern Search, PS) and population-based (Particle Swarm Optimization, PSO2011 and Artificial Bee Colony, ABC) algorithms. A Wilcoxon-Signed Rank Test is performed to measure the pair-wise statistical performances of the algorithms, the results of which indicate that the proposed VS algorithm outperforms the SA, PS and ABC algorithms while being competitive with the PSO2011 algorithm.

A new ECG beat clustering method based on kernelized fuzzy c-means and hybrid ant colony optimization for continuous domains

The kernelized fuzzy c-means algorithm uses kernel methods to improve the clustering performance of the well known fuzzy c-means algorithm by mapping a given dataset into a higher dimensional space non-linearly. Thus, the newly obtained dataset is more likely to be linearly seprable. However, to further improve the clustering performance, an optimization method is required to overcome the drawbacks of the traditional algorithms such as, sensitivity to initialization, trapping into local minima and lack of prior knowledge for optimum paramaters of the kernel functions. In this paper, to overcome these drawbacks, a new clustering method based on kernelized fuzzy c-means algorithm and a recently proposed ant based optimization algorithm, hybrid ant colony optimization for continuous domains, is proposed. The proposed method is applied to a dataset which is obtained from MIT-BIH arrhythmia database. The dataset consists of six types of ECG beats including, Normal Beat (N), Premature Ventricular Contraction (PVC), Fusion of Ventricular and Normal Beat (F), Artrial Premature Beat (A), Right Bundle Branch Block Beat (R) and Fusion of Paced and Normal Beat (f). Four time domain features are extracted for each beat type and training and test sets are formed. After several experiments it is observed that the proposed method outperforms the traditional fuzzy c-means and kernelized fuzzy c-means algorithms.

Analog filter group delay optimization using the Vortex Search algorithm

In this study, a method based on the Vortex Search algorithm is proposed for analog filter group delay optimization. To measure the performance of the proposed method, a number of all-pass filters are first cascaded to a fifth order Chebyshev low-pass filter and then the optimum parameters of these all-pass filters are searched by using the Vortex Search algorithm. It is shown that, the group delay of the fifth order Chebyshev low-pass filter (2.593s) can be decreased down to 1.547s by using an optimum fourth order all-pass filter.

Performance evaluation of Radial Basis Function Neural Network on ECG beat classification

In this study Radial Basis Function Neural Network (RBFNN) was trained by different methods to study performance of each method on classification of ECG beats. To train the neural networks six types of beats including, Normal Beat (N), Premature Ventricular Contraction (PVC), Fusion of Ventricular and Normal Beat (F), Atrial Premature Beat (A), Right Bundle Branch Blok Beat (R), and Fusion of Paced and Normal Beat (f) were selected from the MIT-BIH arrhythmia database. Training of the neural networks were performed with a training set which includes 100 beats for each class. Four time domain (morphological) features were extracted from the beats for classification process. Then several experiments were performed over the test set, and it was observed that, combining RBFNN with different methods has a positive effect on the classification performance of ECG beats.

A novel state space representation for the solution of 2D-HP protein folding problem using reinforcement learning methods

A new state space representation of the protein folding problem in 2D-HP model is proposed for the use of reinforcement learning methods.The proposed state space representation reduces the dependency of the size of the state-action space to the amino acid sequence length.The proposed state space representation also provides an actual learning for an agent. Thus, at the end of a learning process an agent could find the optimum fold of any sequence of a certain length, which is not the case in the existing reinforcement learning methods.By using the Ant-Q algorithm (an ant based reinforcement learning method), optimum fold of a protein sequence is found rapidly when compared to the standard Q-learning algorithm. In this study, a new state space representation of the protein folding problem for the use of reinforcement learning methods is proposed. In the existing studies, the way of defining the state-action space prevents the agent to learn the state space for any amino-acid sequence, but rather, the defined state-action space is valid for only a particular amino-acid sequence. Moreover, in the existing methods, the size of the state space is strictly depends on the amino-acid sequence length. The newly proposed state-action space reduces this dependency and allows the agent to find the optimal fold of any sequence of a certain length. Additionally, by utilizing an ant based reinforcement learning algorithm, the Ant-Q algorithm, optimum fold of a protein is found rapidly when compared to the standard Q-learning algorithm. Experiments showed that, the new state-action space with the ant based reinforcement learning method is much more suited for the protein folding problem in two dimensional lattice model.

ECG beat clustering using fuzzy c-means algorithm and particle swarm optimization

In this paper, an ECG beat clustering method based on fuzzy c-means algorithm and particle swarm optimization is proposed. For this purpose, ECG records which are selected from MIT-BIH arrhythmia database are firstly preprocessed and then four morphological features are extracted for six different types of beats. These features are then clustered with the proposed method. During the classification phase, in order to minimize the incongruity between the experiments and to better evaluate the performance of the proposed system a simple but stable classification method is used. After several experiments it is observed that the proposed method overcomes the restrictions of the fuzzy c-means algorithm which are sensitivity to initialization and trapping into local minima.

Segmentation of brain MRI images by using type-II fuzzy clustering algorithm

In this study, segmentation of Multiple Sclerosis (MS) lesions from synthetic brain MRI images was aimed by using fuzzy clustering algorithms. The performances of fuzzy c-means algorithm and type-2 fuzzy c-means algorithm were compared. After several experiments it was shown that, the type-2 fuzzy c-means algorithm performed better than the standard fuzzy c-means algorithm.