Diego Andina

WBCD breast cancer database classification applying artificial metaplasticity neural network

The correct diagnosis of breast cancer is one of the major problems in the medical field. From the literature it has been found that different pattern recognition techniques can help them to improve in this domain. These techniques can help doctors form a second opinion and make a better diagnosis. In this paper we present a novel improvement in neural network training for pattern classification. The proposed training algorithm is inspired by the biological metaplasticity property of neurons and Shannons information theory. During the training phase the Artificial metaplasticity Multilayer Perceptron (AMMLP) algorithm gives priority to updating the weights for the less frequent activations over the more frequent ones. In this way metaplasticity is modeled artificially. AMMLP achieves a more effcient training, while maintaining MLP performance. To test the proposed algorithm we used the Wisconsin Breast Cancer Database (WBCD). AMMLP performance is tested using classification accuracy, sensitivity and specificity analysis, and confusion matrix. The obtained AMMLP classification accuracy of 99.26%, a very promising result compared to the Backpropagation Algorithm (BPA) and recent classification techniques applied to the same database.

Feature selection using Sequential Forward Selection and classification applying Artificial Metaplasticity Neural Network

The feature selection has been widely used to reduce the data dimensionality. Data reduction improve the classification performance, the approximation function, and pattern recognition systems in terms of speed, accuracy and simplicity. A strategy to reduce the number of features in local search are the sequential search algorithms. In this work is presented a feature selection method based on Sequential Forward Selection (SFS) and Feed Forward Neural Network (FFNN) to estimate the prediction error as a selection criterion. Three well-known database have been used to test the SFS-FFNN with Artificial Metaplasticity on Perceptron Multilayer (AMMLP). The AMMLP is a new method applied for classification of patterns. The results obtained by SFS-FFNN with AMMLP in classification accuracy are superior than obtained by conventional BP algorithm and other recent feature selection algorithms applied to the same database. By these reasons the proposed method SFS-FFNN with AMMLP is an interesting alternative to reduce the data dimensionality and provide a high accuracy.

Breast cancer classification applying artificial metaplasticity algorithm

A novel improvement in neural network training for pattern classification is presented in this paper. The proposed training algorithm is inspired by the biological metaplasticity property of neurons and Shannons information theory. This algorithm is applicable to artificial neural networks (ANNs) in general, although here it is applied to a multilayer perceptron (MLP). During the training phase, the artificial metaplasticity multilayer perceptron (AMMLP) algorithm assigns higher values for updating the weights in the less frequent activations than in the more frequent ones. AMMLP achieves a more efficient training and improves MLP performance. The well-known and readily available Wisconsin Breast Cancer Database (WBCD) has been used to test the algorithm. Performance of the AMMLP was tested through classification accuracy, sensitivity and specificity analysis, and confusion matrix analysis. The results obtained by AMMLP are compared with the backpropagation algorithm (BPA) and other recent classification techniques applied to the same database. The best result obtained so far with the AMMLP algorithm is 99.63%.

Ant based edge linking algorithm

Conventional image edge detectors always result in missing parts of the edges. Broken edge linking is an image improvement technique that is complementary to edge detection, where the broken edges are connected to form closed contours in order to separate the regions in the image. In this paper, Ant System (AS) algorithm is modified for edge linking problem. As input, a binary image obtained after applying the Sobel edge operator is used. The proposed method defines a novel fitness function dependent on two variables: the grayscale visibility of the pixels and the length of the connecting edge, in order to obtain effective solution evaluation. Another novelty is of applying the grayscale visibility matrix as the initial pheromone trails matrix so that the pixels belonging to true edges have a higher probability of being chosen by ants on their initial routes, which reduces computational load. The results of the experiments are presented to confirm the effectiveness of the proposed method.

Edge detection using ant colony search algorithm and multiscale contrast enhancement

In this paper, Ant Colony System (ACS) algorithm is applied for edge detection in grayscale images. The novelty of the proposed method is to extract a set of images from the original grayscale image using Multiscale Adaptive Gain for image contrast enhancement and then apply the ACS algorithm to detect the edges on each of the extracted images. The resulting set of images represents the pheromone trails matrices which are summed to produce the output image. The image contrast enhancement makes ACS algorithm more effective when accumulating pheromone trails on the true edge pixels. The results of the experiments are presented to confirm the effectiveness of the proposed method.

Performance Analysis of Neural Network Detectors by Importance Sampling Techniques

Often, Neural Networks are involved in binary detectors of communication, radar or sonar systems. The design phase of a neural network detector usually requires the application of Monte Carlo trials in order to estimate some performance parameters.The classical Monte Carlo method is suitable to estimate high event probabilities (higher than 0.01), but not suitable to estimate very low event probabilities (say, 10−5 or less). For estimations of very low false alarm probabilities (or error probabilities), a modified Monte Carlo technique, the so-called Importance Sampling (IS) technique, is considered in this paper; some topics are developed, such as optimal and suboptimal IS probability density functions (biasing density functions), control parameters and new algorithms for the minimization of the estimator error.The main novelty of this paper is the application of an efficient IS technique on neural networks, drastically reducing the number of patterns required for testing events of low probability. As a practical application, the IS technique is applied to a neural detector on a radar (or sonar) system.

Wood defects classification using Artificial Metaplasticity neural network

Artificial Metaplasticity (AMP) is a novel Artificial Neural Network (ANN) training algorithm inspired in biological metaplasticity property of neurons and Shannons information theory. During training phase, the AMP training algorithm gives more relevance to the less frequent patterns and subtracts relevance to the frequent ones, achieving a much more efficient training, while at least maintaining the MLPs performance. AMP is specially recommended when few patterns are available to train the network. In this paper, we implement an Artificial Metaplasticity MLP (AMMLP) in order to classify defects in wood images. The defects are three different types of knots found in wood surfaces. Classification is based on the features obtained from Gabor filters. Experimental results show that AMMLPs reach better accuracy than the classical BP algorithm as well as with recently proposed algorithms applied on the same database.

Feature Vectors Generation for Detection of Microcalcifications in Digitized Mammography Using Neural Networks

This paper presents and tests a methodology that sinergically combines a select of successful advances in each step to automatically classify microcalcifications (MCs) in digitized mammography. The method combines selection of regions of interest (ROI), enhancement by histogram adaptive techniques, processing by multiscale wavelet and gray level statistical techniques, generation, clustering and labelling of suboptimal feature vectors (SFVs), and a Neural feature selector and detector to finally classify the MCs. The experimental results with the method promise interesting advances in the problem of automatic detection and classification of MCs1.

Images sub-segmentation with the PFCM clustering algorithm

In this work we propose a method for sub-segmentation of images using the PFCM clustering algorithm. The sub-segmentation consists of finding, within the clusters found using the segmentation process, those data less representative, or atypical data, belonging to the clusters. These data represent, in many cases, the zones of interest during image analysis. Two different examples are used in order to show the results, and the advantages of identifying those elements of data forced to belong to a cluster, of which they are the less representative and, therefore may contain information of great interest in particular applications.

Comparison of a neural network detector vs Neyman-Pearson optimal detector

We optimize a neural network applied to binary detection such as those found in radar or sonar. Topics about designing the structure, training procedure and evaluating the performance, are discussed. The detector optimization is based on the use of a criterion function that yields a solution significantly superior to the typical sum-of-square-error. Using a modeled input, its performance is evaluated by Monte Carlo trials. As a result, detection curves are compared with the theoretical optimum ones (Neyman-Pearson detectors). For the model, and despite of the blind learning of the neural network, its performance is very close to optimal.