Clarimar José Coelho

Comparison of extrasystolic ECG signal classifiers using discrete wavelet transforms

This work compares and contrasts results of classifying time-domain ECG signals with pathological conditions taken from the MIT-BIH arrhythmia database. Linear discriminant analysis and a multi-layer perceptron were used as classifiers. The neural network was trained by two different methods, namely back-propagation and a genetic algorithm. Converting the time-domain signal into the wavelet domain reduced the dimensionality of the problem at least 10-fold. This was achieved using wavelets from the db6 family as well as using adaptive wavelets generated using two different strategies. The wavelet transforms used in this study were limited to two decomposition levels. A neural network with evolved weights proved to be the best classifier with a maximum of 99.6% accuracy when optimised wavelet-transform ECG data was presented to its input and 95.9% accuracy when the signals presented to its input were decomposed using db6 wavelets. The linear discriminant analysis achieved a maximum classification accuracy of 95.7% when presented with optimised and 95.5% with db6 wavelet coefficients. It is shown that the much simpler signal representation of a few wavelet coefficients obtained through an optimised discrete wavelet transform facilitates the classification of non-stationary time-variant signals task considerably. In addition, the results indicate that wavelet optimisation may improve the classification ability of a neural network.

A solution to the wavelet transform optimization problem in multicomponent analysis

The wavelet transform has been shown to be an efficient tool for data treatment in multivariate calibration. However, previous works had the limitation of using fixed wavelets, which must be chosen a priori, because adjusting the wavelets to the data set involves a complex constrained optimization problem. This difficulty is overcome here and the mathematical background involved is described in detail. The proposed approach maximizes the compression performance of the quadrature-mirror filter bank used to process the spectra. After the optimization phase, the recently proposed successive projections algorithm is used to select subsets of wavelet coefficients in order to minimize collinearity problems in the regression. To demonstrate the efficiency of the entire strategy, a low-resolution ICP-AES was deliberately chosen to tackle a hard multivariate calibration problem involving the simultaneous multicomponent determination of Mn, Mo, Cr, Ni and Fe in steel samples. This analysis is intrinsically complex, due to strong collinearity and severe spectral overlapping, problems that are aggravated by the use of low-resolution optics. Moreover, there are also several regions in the spectra where the signal-to-noise ratio is poor. The results demonstrate that the optimization leads to models with better parsimony and prediction ability when compared to the fixed-wavelet approach adopted in previous papers.

Optimal discrimination and classification of THz spectra in the wavelet domain.

In rapid scan Fourier transform spectrometry, we show that the noise in the wavelet coefficients resulting from the filter bank decomposition of the complex insertion loss function is linearly related to the noise power in the sample interferogram by a noise amplification factor. By maximizing an objective function composed of the power of the wavelet coefficients divided by the noise amplification factor, optimal feature extraction in the wavelet domain is performed. The performance of a classifier based on the output of a filter bank is shown to be considerably better than that of an Euclidean distance classifier in the original spectral domain. An optimization procedure results in a further improvement of the wavelet classifier. The procedure is suitable for enhancing the contrast or classifying spectra acquired by either continuous wave or THz transient spectrometers as well as for increasing the dynamic range of THz imaging systems.

A linear semi-infinite programming strategy for constructing optimal wavelet transforms in multivariate calibration problems

A novel strategy for the optimization of wavelet transforms with respect to the statistics of the data set in multivariate calibration problems is proposed. The optimization follows a linear semi-infinite programming formulation, which does not display local maxima problems and can be reproducibly solved with modest computational effort. After the optimization, a variable selection algorithm is employed to choose a subset of wavelet coefficients with minimal collinearity. The selection allows the building of a calibration model by direct multiple linear regression on the wavelet coefficients. In an illustrative application involving the simultaneous determination of Mn, Mo, Cr, Ni, and Fe in steel samples by ICP-AES, the proposed strategy yielded more accurate predictions than PCR, PLS, and nonoptimized wavelet regression.

A GPU-Based Implementation of the Firefly Algorithm for Variable Selection in Multivariate Calibration Problems

Several variable selection algorithms in multivariate calibration can be accelerated using Graphics Processing Units (GPU). Among these algorithms, the Firefly Algorithm (FA) is a recent proposed metaheuristic that may be used for variable selection. This paper presents a GPU-based FA (FA-MLR) with multiobjective formulation for variable selection in multivariate calibration problems and compares it with some traditional sequential algorithms in the literature. The advantage of the proposed implementation is demonstrated in an example involving a relatively large number of variables. The results showed that the FA-MLR, in comparison with the traditional algorithms is a more suitable choice and a relevant contribution for the variable selection problem. Additionally, the results also demonstrated that the FA-MLR performed in a GPU can be five times faster than its sequential implementation.

Parallelization of a Modified Firefly Algorithm using GPU for Variable Selection in a Multivariate Calibration Problem

The recent improvements of Graphics Processing Units (GPU) have provided to the bio-inspired algorithms a powerful processing platform. Indeed, a lot of highly parallelizable problems can be significantly accelerated using GPU architecture. Among these algorithms, the Firefly Algorithm (FA) is a newly proposed method with potential application in several real world problems such as variable selection problem in multivariate calibration. The main drawback of this task lies in its computation burden, as it grows polynomially with the number of variables available. In this context, this paper proposes a GPU-based FA for variable selection in a multivariate calibration problem. Such implementation is aimed at improving the computational efficiency of the algorithm. For this purpose, a new strategy of regression coefficients calculation is employed. The advantage of the proposed implementation is demonstrated in an example involving a large number of variables. In such example, gains of speedup were obtained. Additionally the authors also demonstrate that the FA, in comparison with traditional algorithms, can be a relevant contribution for the variable selection problem.

Multi-objective evolutionary algorithm for variable selection in calibration problems: A case study for protein concentration prediction

This paper presents a multi-objective formulation for variable selection in calibration problems. The prediction of protein concentration on wheat is obtained by a linear regression model using variables obtained by a spectrophotometer device. This device measure hundreds of correlated variables related with physicochemical properties and that can be used to estimate the protein concentration. The problem is the selection of a subset informative and uncorrelated variables that help the minimization of prediction error. In this work we propose the use of two objectives in this problem: the prediction error and the number of variables in the model, both related to linear equations system stability. We proposed a multi-objective formulation using two multi-objective algorithms: the NSGA-II and the SPEA-II. Additionally we propose a final decision maker method to choice the final subset of variables from the Pareto front. For the case study is used wheat data obtained by NIR spectrometry where the objective is the determination of a variable subgroup with information about protein concentration. The results of traditional techniques of multivariate calibration as the Successive Projections Algorithm (SPA), Partial Least Square (PLS) and mono-objective genetic algorithm are presents for comparisons. For NIR spectral analysis of protein concentration on wheat, the number of variables selected from 775 spectral variables was reduced for just 10 in the SPEA-II algorithm. The prediction error decreased from 0.2 in the classical methods to 0.09 in proposed approach, a reduction of 45%. The model using variables selected by SPEA-II had better prediction performance than classical algorithms and full-spectrum partial least-squares (PLS).

Feature Selection using Genetic Algorithm: An Analysis of the Bias-Property for One-Point Crossover

Genetic algorithms (GAs) have been used for feature selection with binary representation. Even if binary representation has perfect probability to include or remove a feature in the search process, some works in the field of chemometrics have reported criticism about a high number of features selected by GA implementations. Thus, in this paper, we aim to propose an investigation of the number of features selected on a point of view of the bias-property using implementations from the GA-PLS toolboxes (Genetic Algorithm with Partial Least Square). The study is performed using an one-point crossover operator and a common initialization procedure used in the matlab toolboxes. Results show the existence of such a bias that influences the increase in the number of features over the generations.

A feasibility cachaca type recognition using computer vision and pattern recognition

The problem of recognition of aging time and wood type in chacaca is presented.A new approach is introduced using a computer vision system.The developed image capture device and information processing method is presented.Results show that the new technique is cheaper and better than previous approaches. Brazilian rum (also known as cachaca) is the third most commonly consumed distilled alcoholic drink in the world, with approximately 2.5 billion liters produced each year. It is a traditional drink with refined features and a delicate aroma that is produced mainly in Brazil but consumed in many countries. It can be aged in various types of wood for 1-3years, which adds aroma and a distinctive flavor with different characteristics that affect the price. A research challenge is to develop a cheap automatic recognition system that inspects the finished product for the wood type and the aging time of its production. Some classical methods use chemical analysis, but this approach requires relatively expensive laboratory equipment. By contrast, the system proposed in this paper captures image signals from samples and uses an intelligent classification technique to recognize the wood type and the aging time. The classification system uses an ensemble of classifiers obtained from different wavelet decompositions. Each classifier is obtained with different wavelet transform settings. We compared the proposed approach with classical methods based on chemical features. We analyzed 105 samples that had been aged for 3years and we showed that the proposed solution could automatically recognize wood types and the aging time with an accuracy up to 100.00% and 85.71% respectively, and our method is also cheaper.

Multi-objective Genetic Algorithm for Variable Selection in Multivariate Classification Problems

This paper proposes multi-objective genetic algorithm for the problem of variable selection in multivariate calibration. We consider the problem related to the classification of biodiesel samples to detect adulteration, Linear Discriminant Analysis classifier. The goal of the multi--objective algorithm is to reduce the dimensionality of the original set of variables; thus, the classification model can be less sensitive, providing a better generalization capacity. In particular, in this paper we adopted a version of the Non-dominated Sorting Genetic Algorithm (NSGA-II) and compare it to a mono-objective Genetic Algorithm (GA) in terms of sensitivity in the presence of noise. Results show that the mono-objective selects 20 variables on average and presents an error rate of 14%. One the other hand, the multi-objective selects 7 variables and has an error rate of 11%. Consequently, we show that the multi-objective formulation provides classification models with lower sensitivity to the instrumental noise when compared to the mono-objetive formulation