Brain Inspired Cortical Coding Method for Fast Clustering and Codebook Generation.

A major archetype of artificial intelligence is developing algorithms facilitating temporal efficiency and accuracy while boosting the generalization performance. Even with the latest developments in machine learning, a key limitation has been the inefficient feature extraction from the initial data, which is essential in performance optimization. Here, we introduce a feature extraction method inspired by energy-entropy relations of sensory cortical networks in the brain. Dubbed the brain-inspired cortex, the algorithm provides convergence to orthogonal features from streaming signals with superior computational efficiency while processing data in a compressed form. We demonstrate the performance of the new algorithm using artificially created complex data by comparing it with the commonly used traditional clustering algorithms, such as Birch, GMM, and K-means. While the data processing time is significantly reduced-seconds versus hours-encoding distortions remain essentially the same in the new algorithm, providing a basis for better generalization. Although we show herein the superior performance of the cortical coding model in clustering and vector quantization, it also provides potent implementation opportunities for machine learning fundamental components, such as reasoning, anomaly detection and classification in large scope applications, e.g., finance, cybersecurity, and healthcare.

An Efficient Big Data Anonymization Algorithm Based on Chaos and Perturbation Techniques.

The topic of big data has attracted increasing interest in recent years. The emergence of big data leads to new difficulties in terms of protection models used for data privacy, which is of necessity for sharing and processing data. Protecting individuals' sensitive information while maintaining the usability of the data set published is the most important challenge in privacy preserving. In this regard, data anonymization methods are utilized in order to protect data against identity disclosure and linking attacks. In this study, a novel data anonymization algorithm based on chaos and perturbation has been proposed for privacy and utility preserving in big data. The performance of the proposed algorithm is evaluated in terms of Kullback-Leibler divergence, probabilistic anonymity, classification accuracy, F-measure and execution time. The experimental results have shown that the proposed algorithm is efficient and performs better in terms of Kullback-Leibler divergence, classification accuracy and F-measure compared to most of the existing algorithms using the same data set. Resulting from applying chaos to perturb data, such successful algorithm is promising to be used in privacy preserving data mining and data publishing.

Application of canonical correlation analysis for identifying viral integration preferences.

MOTIVATION: Gene therapy aims at using viral vectors for attaching helpful genetic code to target genes. Therefore, it is of great importance to develop methods that can discover significant patterns around viral integration sites. Canonical correlation analysis is an unsupervised statistical tool that is used to describe the relations between two related views of the same semantic object, which fits well for identifying such salient patterns. RESULTS: Proposed method is demonstrated on a sequence dataset obtained from a study on HIV-1 preferred integration regions. The subsequences on the left and right sides of the integration points are given to the method as the two views, and statistically significant relations are found between sequence-driven features derived from these two views, which suggest that the viral preference must be the factor responsible for this correlation. We found that there are significant correlations at x=5 indicating a palindromic behavior surrounding the viral integration site, which complies with the previously reported results. AVAILABILITY: Developed software tool is available at http://ce.istanbul.edu.tr/bioinformatics/hiv1/.

Mammographical mass detection and classification using local seed region growing-spherical wavelet transform (LSRG-SWT) hybrid scheme.

The purpose of this study is to implement accurate methods of detection and classification of benign and malignant breast masses in mammograms. Our new proposed method, which can be used as a diagnostic tool, is denoted Local Seed Region Growing-Spherical Wavelet Transform (LSRG-SWT), and consists of four steps. The first step is homomorphic filtering for enhancement, and the second is detection of the region of interests (ROIs) using a Local Seed Region Growing (LSRG) algorithm, which we developed. The third step incoporates Spherical Wavelet Transform (SWT) and feature extraction. Finally the fourth step is classification, which consists of two sequential components: the 1st classification distinguishes the ROIs as either mass or non-mass and the 2nd classification distinguishes the masses as either benign or malignant using a Support Vector Machine (SVM). The mammograms used in this study were acquired from the hospital of Istanbul University (I.U.) in Turkey and the Mammographic Image Analysis Society (MIAS). The results demonstrate that the proposed scheme LSRG-SWT achieves 96% and 93.59% accuracy in mass/non-mass classification (1st component) and benign/malignant classification (2nd component) respectively when using the I.U. database with k-fold cross validation. The system achieves 94% and 91.67% accuracy in mass/non-mass classification and benign/malignant classification respectively when using the I.U. database as a training set and the MIAS database as a test set with external validation.

Comparison of aggregators for multi-objective SNP selection.

SNPs (Single Nucleotide Polymorphisms) are genomic variants that associate with many genetic characteristics. These variants can also be utilized to track the on-going mutation in population genetics. The goal of this study was to select the most relevant SNP subsets for discriminating ethnic groups. Each SNP was evaluated by its: i) Mutual information, ii) Relief-F score, iii) Loadings of the first principal component, iv) Loadings of the second principal component. Combining these four feature ranking criteria in different ways, three different aggregation methods (Pareto Optimal, Condorcet, MC4) were compared with respect to their SNP selection accuracies. Results showed that SNP subsets chosen with Pareto Optimal yielded better classification accuracy.

Collection and analysis of a Parkinson speech dataset with multiple types of sound recordings.

There has been an increased interest in speech pattern analysis applications of Parkinsonism for building predictive telediagnosis and telemonitoring models. For this purpose, we have collected a wide variety of voice samples, including sustained vowels, words, and sentences compiled from a set of speaking exercises for people with Parkinson's disease. There are two main issues in learning from such a dataset that consists of multiple speech recordings per subject: 1) How predictive these various types, e.g., sustained vowels versus words, of voice samples are in Parkinson's disease (PD) diagnosis? 2) How well the central tendency and dispersion metrics serve as representatives of all sample recordings of a subject? In this paper, investigating our Parkinson dataset using well-known machine learning tools, as reported in the literature, sustained vowels are found to carry more PD-discriminative information. We have also found that rather than using each voice recording of each subject as an independent data sample, representing the samples of a subject with central tendency and dispersion metrics improves generalization of the predictive model.

Selection of vocal features for Parkinson's Disease diagnosis.

Parkinson's Disease (PD) is a neurodegenerative motor system disorder, which also causes vocal impairments for most of its patients. A number of recent exploratory studies have evaluated the feasibility of detecting voice disorders by applying data mining tools to acoustic features extracted from speech recordings of patients. Selection of a minimal yet descriptive set of features is crucial for improving the classifier generalisation capability and interpretability of the classification model as well as for reducing the burden of data preprocessing. We propose a hybrid of feature selection and cross-validation procedures to lower the bias in the assessment of classifier accuracy.

Mammographic mass detection using wavelets as input to neural networks.

The objective of this paper is to demonstrate the utility of artificial neural networks, in combination with wavelet transforms for the detection of mammogram masses as malign or benign. A total of 45 patients who had breast masses in their mammography were enrolled in the study. The neural network was trained on the wavelet based feature vectors extracted from the mammogram masses for both benign and malign data. Therefore, in this study, Multilayer ANN was trained with the Backpropagation, Conjugate Gradient and Levenberg-Marquardt algorithms and ten-fold cross validation procedure was used. A satisfying sensitivity percentage of 89.2% was achieved with Levenberg-Marquardt algorithm. Since, this algorithm combines the best features of the Gauss-Newton technique and the other steepest-descent algorithms and thus it reaches desired results very fast.

A wavelet-based mammographic image denoising and enhancement with homomorphic filtering.

Breast cancer continues to be a significant public health problem in the world. The diagnosing mammography method is the most effective technology for early detection of the breast cancer. However, in some cases, it is difficult for radiologists to detect the typical diagnostic signs, such as masses and microcalcifications on the mammograms. This paper describes a new method for mammographic image enhancement and denoising based on wavelet transform and homomorphic filtering. The mammograms are acquired from the Faculty of Medicine of the University of Akdeniz and the University of Istanbul in Turkey. Firstly wavelet transform of the mammograms is obtained and the approximation coefficients are filtered by homomorphic filter. Then the detail coefficients of the wavelet associated with noise and edges are modeled by Gaussian and Laplacian variables, respectively. The considered coefficients are compressed and enhanced using these variables with a shrinkage function. Finally using a proposed adaptive thresholding the fine details of the mammograms are retained and the noise is suppressed. The preliminary results of our work indicate that this method provides much more visibility for the suspicious regions.