Global and local detection of liver steatosis from ultrasound

Liver steatosis is a common disease usually associated with social and genetic factors. Early detection and quantification is important since it can evolve to cirrhosis. Steatosis is usually a diffuse liver disease, since it is globally affected. However, steatosis can also be focal affecting only some foci difficult to discriminate. In both cases, steatosis is detected by laboratorial analysis and visual inspection of ultrasound images of the hepatic parenchyma. Liver biopsy is the most accurate diagnostic method but its invasive nature suggest the use of other non-invasive methods, while visual inspection of the ultrasound images is subjective and prone to error. In this paper a new Computer Aided Diagnosis (CAD) system for steatosis classification and analysis is presented, where the Bayes Factor, obatined from objective intensity and textural features extracted from US images of the liver, is computed in a local or global basis. The main goal is to provide the physician with an application to make it faster and accurate the diagnosis and quantification of steatosis, namely in a screening approach. The results showed an overall accuracy of 93.54% with a sensibility of 95.83% and 85.71% for normal and steatosis class, respectively. The proposed CAD system seemed suitable as a graphical display for steatosis classification and comparison with some of the most recent works in the literature is also presented.

Chronic liver disease staging classification based on ultrasound, clinical and laboratorial data

In this work the identification and diagnosis of various stages of chronic liver disease is addressed. The classification results of a support vector machine, a decision tree and a k-nearest neighbor classifier are compared. Ultrasound image intensity and textural features are jointly used with clinical and laboratorial data in the staging process. The classifiers training is performed by using a population of 97 patients at six different stages of chronic liver disease and a leave-one-out cross-validation strategy. The best results are obtained using the support vector machine with a radial-basis kernel, with 73.20% of overall accuracy. The good performance of the method is a promising indicator that it can be used, in a non invasive way, to provide reliable information about the chronic liver disease staging.

Avaliação da caracterização de lesões em mamografia com recurso a sistemas CAD (Computer-Aided Diagnosis)

Os sistemas Computer-Aided Diagnosis (CAD) auxiliam a deteção e diferenciação de lesões benignas e malignas, aumentando a performance no diagnóstico do cancro da mama. As lesões da mama estão fortemente correlacionadas com a forma do contorno: lesões benignas apresentam contornos regulares, enquanto as lesões malignas tendem a apresentar contornos irregulares. Desta forma, a utilização de medidas quantitativas, como a dimensão fractal (DF), pode ajudar na caracterização dos contornos regulares ou irregulares de uma lesão. O principal objetivo deste estudo é verificar se a utilização concomitante de 2 (ou mais) medidas de DF – uma tradicionalmente utilizada, a qual foi designada por “DF de contorno”; outra proposta por nós, designada por “DF de área” – e ainda 3 medidas obtidas a partir destas, por operações de dilatação/erosão e por normalização de uma das medidas anteriores, melhoram a capacidade de caracterização de acordo com a escala BIRADS (Breast Imaging Reporting and Data System) e o tipo de lesão. As medidas de DF (DF contorno e DF área) foram calculadas através da aplicação do método box-counting, diretamente em imagens de lesões segmentadas e após a aplicação de um algoritmo de dilatação/erosão. A última medida baseia-se na diferença normalizada entre as duas medidas DF de área antes e após a aplicação do algoritmo de dilatação/erosão. Os resultados demonstram que a medida DF de contorno é uma ferramenta útil na diferenciação de lesões, de acordo com a escala BIRADS e o tipo de lesão; no entanto, em algumas situações, ocorrem alguns erros. O uso combinado desta medida com as quatro medidas propostas pode melhorar a classificação das lesões.

Dominant set approach to ECG biometrics

Electrocardiographic (ECG) signals are emerging as a recent trend in the field of biometrics. In this paper, we propose a novel ECG biometric system that combines clustering and classification methodologies. Our approach is based on dominant-set clustering, and provides a framework for outlier removal and template selection. It enhances the typical workflows, by making them better suited to new ECG acquisition paradigms that use fingers or hand palms, which lead to signals with lower signal to noise ratio, and more prone to noise artifacts. Preliminary results show the potential of the approach, helping to further validate the highly usable setups and ECG signals as a complementary biometric modality.

Automatic diagnosis of liver steatosis by ultrasound using autoregressive tissue characterization

Liver steatosis is mainly a textural abnormality of the hepatic parenchyma due to fat accumulation on the hepatic vesicles. Today, the assessment is subjectively performed by visual inspection. Here a classifier based on features extracted from ultrasound (US) images is described for the automatic diagnostic of this phatology. The proposed algorithm estimates the original ultrasound radio-frequency (RF) envelope signal from which the noiseless anatomic information and the textural information encoded in the speckle noise is extracted. The features characterizing the textural information are the coefficients of the first order autoregressive model that describes the speckle field. A binary Bayesian classifier was implemented and the Bayes factor was calculated. The classification has revealed an overall accuracy of 100%. The Bayes factor could be helpful in the graphical display of the quantitative results for diagnosis purposes.

Determining the number of clusters in categorical data

Cluster analysis for categorical data has been an active area of research. A well-known problem in this area is the determination of the number of clusters, which is unknown and must be inferred from the data. In order to estimate the number of clusters, one often resorts to information criteria, such as BIC (Bayesian information criterion), MML (minimum message length, proposed by Wallace and Boulton, 1968), and ICL (integrated classification likelihood). In this work, we adopt the approach developed by Figueiredo and Jain (2002) for clustering continuous data. They use an MML criterion to select the number of clusters and a variant of the EM algorithm to estimate the model parameters. This EM variant seamlessly integrates model estimation and selection in a single algorithm. For clustering categorical data, we assume a finite mixture of multinomial distributions and implement a new EM algorithm, following a previous version (Silvestre et al., 2008). Results obtained with synthetic datasets are encouraging. The main advantage of the proposed approach, when compared to the above referred criteria, is the speed of execution, which is especially relevant when dealing with large data sets.

Probabilistic consensus clustering using evidence accumulation

Clustering ensemble methods produce a consensus partition of a set of data points by combining the results of a collection of base clustering algorithms. In the evidence accumulation clustering (EAC) paradigm, the clustering ensemble is transformed into a pairwise co-association matrix, thus avoiding the label correspondence problem, which is intrinsic to other clustering ensemble schemes. In this paper, we propose a consensus clustering approach based on the EAC paradigm, which is not limited to crisp partitions and fully exploits the nature of the co-association matrix. Our solution determines probabilistic assignments of data points to clusters by minimizing a Bregman divergence between the observed co-association frequencies and the corresponding co-occurrence probabilities expressed as functions of the unknown assignments. We additionally propose an optimization algorithm to find a solution under any double-convex Bregman divergence. Experiments on both synthetic and real benchmark data show the effectiveness of the proposed approach.

Partial classification of Lorenz knots: Syllable permutations of torus knots words

We define families of aperiodic words associated to Lorenz knots that arise naturally as syllable permutations of symbolic words corresponding to torus knots. An algorithm to construct symbolic words of satellite Lorenz knots is defined. We prove, subject to the validity of a previous conjecture, that Lorenz knots coded by some of these families of words are hyperbolic, by showing that they are neither satellites nor torus knots and making use of Thurston's theorem. Infinite families of hyperbolic Lorenz knots are generated in this way, to our knowledge, for the first time. The techniques used can be generalized to study other families of Lorenz knots.

A MAP approach to evidence accumulation clustering

The Evidence Accumulation Clustering (EAC) paradigm is a clustering ensemble method which derives a consensus partition from a collection of base clusterings obtained using different algorithms. It collects from the partitions in the ensemble a set of pairwise observations about the co-occurrence of objects in a same cluster and it uses these co-occurrence statistics to derive a similarity matrix, referred to as co-association matrix. The Probabilistic Evidence Accumulation for Clustering Ensembles (PEACE) algorithm is a principled approach for the extraction of a consensus clustering from the observations encoded in the co-association matrix based on a probabilistic model for the co-association matrix parameterized by the unknown assignments of objects to clusters. In this paper we extend the PEACE algorithm by deriving a consensus solution according to a MAP approach with Dirichlet priors defined for the unknown probabilistic cluster assignments. In particular, we study the positive regularization effect of Dirichlet priors on the final consensus solution with both synthetic and real benchmark data.

Text classification using compression-based dissimilarity measures

Arguably, the most difficult task in text classification is to choose an appropriate set of features that allows machine learning algorithms to provide accurate classification. Most state-of-the-art techniques for this task involve careful feature engineering and a pre-processing stage, which may be too expensive in the emerging context of massive collections of electronic texts. In this paper, we propose efficient methods for text classification based on information-theoretic dissimilarity measures, which are used to define dissimilarity-based representations. These methods dispense with any feature design or engineering, by mapping texts into a feature space using universal dissimilarity measures; in this space, classical classifiers (e.g. nearest neighbor or support vector machines) can then be used. The reported experimental evaluation of the proposed methods, on sentiment polarity analysis and authorship attribution problems, reveals that it approximates, sometimes even outperforms previous state-of-the-art techniques, despite being much simpler, in the sense that they do not require any text pre-processing or feature engineering.

Comparing clustering solutions: the use of adjusted paired indices

In the present paper we compare clustering solutions using indices of paired agreement. We propose a new method - IADJUST - to correct indices of paired agreement, excluding agreement by chance. This new method overcomes previous limitations known in the literature as it permits the correction of any index. We illustrate its use in external clustering validation, to measure the accordance between clusters and an a priori known structure. The adjusted indices are intended to provide a realistic measure of clustering performance that excludes agreement by chance with ground truth. We use simulated data sets, under a range of scenarios - considering diverse numbers of clusters, clusters overlaps and balances - to discuss the pertinence and the precision of our proposal. Precision is established based on comparisons with the analytical approach for correction specific indices that can be corrected in this way are used for this purpose. The pertinence of the proposed correction is discussed when making a detailed comparison between the performance of two classical clustering approaches, namely Expectation-Maximization (EM) and K-Means (KM) algorithms. Eight indices of paired agreement are studied and new corrected indices are obtained.