Abdel-Badeeh M. Salem

Hybrid intelligent techniques for MRI brain images classification

This paper presents a hybrid technique for the classification of the magnetic resonance images (MRI). The proposed hybrid technique consists of three stages, namely, feature extraction, dimensionality reduction, and classification. In the first stage, we have obtained the features related to MRI images using discrete wavelet transformation (DWT). In the second stage, the features of magnetic resonance images have been reduced, using principal component analysis (PCA), to the more essential features. In the classification stage, two classifiers have been developed. The first classifier based on feed forward back-propagation artificial neural network (FP-ANN) and the second classifier is based on k-nearest neighbor (k-NN). The classifiers have been used to classify subjects as normal or abnormal MRI human images. A classification with a success of 97% and 98% has been obtained by FP-ANN and k-NN, respectively. This result shows that the proposed technique is robust and effective compared with other recent work.

Computer-aided diagnosis of human brain tumor through MRI: A survey and a new algorithm

Computer-aided detection/diagnosis (CAD) systems can enhance the diagnostic capabilities of physicians and reduce the time required for accurate diagnosis. The objective of this paper is to review the recent published segmentation and classification techniques and their state-of-the-art for the human brain magnetic resonance images (MRI). The review reveals the CAD systems of human brain MRI images are still an open problem. In the light of this review we proposed a hybrid intelligent machine learning technique for computer-aided detection system for automatic detection of brain tumor through magnetic resonance images. The proposed technique is based on the following computational methods; the feedback pulse-coupled neural network for image segmentation, the discrete wavelet transform for features extraction, the principal component analysis for reducing the dimensionality of the wavelet coefficients, and the feed forward back-propagation neural network to classify inputs into normal or abnormal. The experiments were carried out on 101 images consisting of 14 normal and 87 abnormal (malignant and benign tumors) from a real human brain MRI dataset. The classification accuracy on both training and test images is 99% which was significantly good. Moreover, the proposed technique demonstrates its effectiveness compared with the other machine learning recently published techniques. The results revealed that the proposed hybrid approach is accurate and fast and robust. Finally, possible future directions are suggested.

Fiducial feature reduction analysis for electrocardiogram (ECG) based biometric recognition

Although the electrocardiogram (ECG) has been a reliable diagnostic tool for decades, its deployment in the context of biometrics is relatively recent. Its robustness to falsification, the evidence it carries about aliveness and its rich feature space has rendered the deployment of ECG based biometrics an interesting prospect. The rich feature space contains fiducial based information such as characteristic peaks which reflect the underlying physiological properties of the heart. The principal goal of this study is to quantitatively evaluate the information content of the fiducial based feature set in terms of their effect on subject and heart beat classification accuracy (ECG data acquired from the PhysioNet ECG repository). To this end, a comprehensive set of fiducial based features was extracted from a collection of ECG records. This feature set was subsequently reduced using a variety of feature extraction/selection methods such as principle component analysis (PCA), linear discriminant analysis (LDA), information-gain ratio (IGR), and rough sets (in conjunction with the PASH algorithm). The performance of the reduced feature set was examined and the results evaluated with respect to the full feature set in terms of the overall classification accuracy and false (acceptance/rejection) ratios (FAR/FRR). The results of this study indicate that the PASH algorithm, deployed within the context of rough sets, reduced the dimensionality of the feature space maximally, while maintaining maximal classification accuracy.

Advanced Machine Learning Technologies and Applications

The recognition of a character begins with analyzing its form and extracting the features that will be exploited for the identification. Primitives can be described as a tool to distinguish an object of one class from another object of another class. It is necessary to define the significant primitives. The size of vector primitives can be large if a large number of primitives are extracted including redundant and irrelevant features. As a result, the performance of the recognition system becomes poor, and as the number of features increases, so does the computing time. Feature selection, therefore, is required to ensure the selection of a subset of features that gives accurate recognition. In our work we propose a feature selection approach based genetic algorithm to improve the discrimination capacity of the Multilayer Perceptron Neural Networks (MLP).

A breast cancer classifier based on a combination of case-based reasoning and ontology approach

Breast cancer is the second most common form of cancer amongst females and also the fifth most cause of cancer deaths worldwide. In case of this particular type of malignancy, early detection is the best form of cure and hence timely and accurate diagnosis of the tumor is extremely vital. Extensive research has been carried out on automating the critical diagnosis procedure as various machine learning algorithms have been developed to aid physicians in optimizing the decision task effectively. In this research, we present a benign/malignant breast cancer classification model based on a combination of ontology and case-based reasoning to effectively classify breast cancer tumors as either malignant or benign. This classification system makes use of clinical data. Two CBR object-oriented frameworks based on ontology are used jCOLIBRI and myCBR. A breast cancer diagnostic prototype is built. During prototyping, we examine the use and functionality of the two focused frameworks.

A wavelet feature extraction method for electrocardiogram (ECG)-based biometric recognition

This paper proposes a discrete wavelet feature extraction method for an electrocardiogram (ECG)-based biometric system. In this method, the RR intervals are extracted and decomposed using discrete biorthogonal wavelet in wavelet coefficient structures. These structures are reduced by excluding the non-informative coefficients, and then, they are fed into a radial basis functions (RBF) neural network for classification. Moreover, the ability of using only the QT or QRS intervals instead of the RR intervals is also investigated. Finally, the results achieved by our method outperformed the auto-correlation (AC)/discrete cosine transform (DCT) method where the DCT coefficients are derived from the AC of ECG segments and fed into the RBF network for classification. The conducted experiments were validated using four Physionet databases. Critical issues like stability overtime, the ability to reject impostors, scalability and generalization to other datasets have also been addressed.

Using data mining for assessing diagnosis of breast cancer

The capability of the classification SVM, Tree Boost and Tree Forest in analyzing the DDSM dataset was investigated for the extraction of the mammographic mass features along with age that discriminates true and false cases. In the present study, SVM technique shows promising results for increasing diagnostic accuracy of classifying the cases witnessed by the largest area under the ROC curve (area under empirical ROC curve =0.79768 and area under binomial ROC curve = 0.85323) comparable to empirical ROC and binomial ROC of 0.57575 and 0.58548 for tree forest while least empirical ROC and binomial ROC of 0.53452 and 0.53882 was accounted by tree boost. These results are confirmed by SVM average gain of 1.7323, tree forest average gain of 1.5576 and tree boost average gain of 1.5718.

Machine learning in electrocardiogram diagnosis

The electrocardiogram (ECG) is a measure of the electrical activity of the heart. Since its introduction in 1887 by Waller, it has been used as a clinical tool for evaluating heart function. A number of cardiovascular diseases (CVDs) (arrhythmia, atrial fibrillation, atrioventricular (AV) dysfunctions, and coronary arterial disease, etc.) can be detected non-invasively using ECG monitoring devices. With the advent of modern signal processing and machine learning techniques, the diagnostic power of the ECG has expanded exponentially. The principal reason for this is the expanded set of features that are typically extracted from the ECG time series. The enhanced feature space provides a wide range of attributes that can be employed in a variety of machine learning techniques, with the goal of providing tools to assist in CVD classification. This paper summarizes some of the principle machine learning approaches to ECG classification, evaluating them in terms of the features they employ, the type(s) of CVD(s) to which they are applied, and their classification accuracy.

Strictness Petroleum Prediction System Based on Fuzzy Model

Petroleum exploration and production is an industry that provides researchers with multi-variant challenging “real world†properties. Recently, some petroleum soft computing techniques have gained a greater interest in prediction within the oil industry. This paper is interested in the analysis, classifying, mining and predictions, based on fuzzy as an intelligent system and an intelligent system called the Strictness Petroleum Prediction System (SPPS), predicted results and statues of crude oil wells and they are compared with other measurement petroleum values. The evaluation study applies test cases, regression models and time series forecasting of vague petroleum datasets to achieve more accurate results. A regression model was made to show the effect of re-testing the prediction processes of petroleum factors. Prediction in time series using a non-parametric functional technique is considered, based on data which was collected from different sources (Daqing oilfield in China and distinct oilfields in Yemen).

Tridiurnal variations in cosmic-ray intensity

Abstract The tridiurnal wave in cosmic-ray intensity expected from a free space anisotropy is theoretically calculated for different cosmic-ray stations which are characterized by different shapes of asymptotic cones of acceptance. The amplitude A and the time of maximum Tmax are given for latitude dependence of the form cosn λ and rigidity dependence of the form R β exp (−(R− 1 R 0 )) , where λ and R are the latitude and rigidity respectively and n, β, R0 are constants. The values of A and Tmax, are calculated for different values of n, β and R0 for each station. The dependence of A and Tmax on the anisotropy parameters is studied for the proper selection of cosmic-ray stations whose data may be used in determining these parameters. Available experimental data were used to find the observed amplitudes of the tridiurnal variations at five stations using power spectrum analysis with hanning applied on the averaged trains. Minimum variance analysis of the theoretical and experimental amplitudes showed that β has a value between 1 and 2, R0 greater than 100 GV and n smaller than 3.