http://hdl.handle.net/2117/3691 2013-05-24T11:58:26Z http://hdl.handle.net/2117/17987 Title: Intelligent management of sepsis in the intensive care unit Authors: Ribas Ripoll, Vicent; Ruiz Rodríguez, Juan Carlos; Vellido Alcacena, Alfredo 2013-02-26T15:01:43Z http://hdl.handle.net/2117/17986 Title: On the use of graphical models to study ICU outcome prediction in septic patients treated with statins Authors: Ribas, Vicent J.; Caballero López, Jesús; Sáez de Tejada, Anna; Ruiz Rodríguez, Juan Carlos; Ruiz Sanmartin, Adolfo; Rello, Jordi; Vellido Alcacena, Alfredo 2013-02-26T14:24:00Z http://hdl.handle.net/2117/17985 Title: Complementing kernel-based visualization of protein sequences with their phylogenetic tree Authors: Cárdenas, Martha Ivón; Vellido Alcacena, Alfredo; Olier, Iván; Rovira, Xavier; Giraldo, Jesus 2013-02-26T14:17:25Z http://hdl.handle.net/2117/17984 Title: Discovering hidden pathways in bioinformatics Authors: Lisboa, Paulo J.G.; Jarman, Ian H.; Etchells, Terence A.; Chambers, Simon J.; Bacciu, Davide; Whittaker, Joe; Garibaldi, Jon M.; Ortega Martorell, Sandra; Vellido Alcacena, Alfredo; Ellis, Ian O. 2013-02-26T14:08:36Z http://hdl.handle.net/2117/17939 Title: Statistical approaches for modeling in microbial source tracking Authors: Belanche Muñoz, Luis Antonio; Blanch, Anicet R. Abstract: Microbial source tracking (MST) concerns the definition of new indicators and appropriate detection methods, the identification of host-specific indicators of fecal pollution, and ultimately the development of useful and reliable predictive models for practical deployment. Optimal predictive models should be designed using proper statistical and computational tools for the analysis of the available data samples. A further requirement is found in the determination of appropriate sets of predictors (indicators, tracers) for developing accurate and low-cost MST solutions. This chapter briefly reviews some of these modeling tools, and their use and feasibility in providing more accurate MST-based results. It also evaluates the potential of established and new algorithmic methods to the identification of fecal pollution sources. 2013-02-22T12:56:43Z http://hdl.handle.net/2117/17935 Title: Learning with heterogeneous neural networks Authors: Belanche Muñoz, Luis Antonio Abstract: This chapter studies a class of neuron models that computes a user-defined similarity function between inputs and weights. The neuron transfer function is formed by composition of an adapted logistic function with the quasi-linear mean of the partial input-weight similarities. The neuron model is capable of dealing directly with mixtures of continuous as well as discrete quantities, among other data types and there is provision for missing values. An artificial neural network using these neuron models is trained using a breeder genetic algorithm until convergence. A number of experiments are carried out in several real-world problems in very different application domains described by mixtures of variales of distinct types and eventually showing missing values. This heterogeneous network is compared to a standard radial basis function network and to a multi-layer perceptron networks and shown to learn from with superior generalization ability at a comparable computational cost. A further important advantage of the resulting neural solutions is the great interpretability of the learned weights, which is done in terms of weighted similarities to prototypes. 2013-02-22T12:18:30Z http://hdl.handle.net/2117/17284 Title: Preprocessing MRS information for classification of human brain tumours Authors: Arizmendi Pereira, Carlos Julio; Vellido Alcacena, Alfredo; Romero Merino, Enrique 2013-01-11T15:18:55Z http://hdl.handle.net/2117/17282 Title: Kernel generative topographic mapping of protein sequences Authors: Cárdenas, Martha Ivón; Vellido Alcacena, Alfredo; Olier Caparroso, Iván; Rovira, Xavier; Giraldo, Jesús 2013-01-11T14:57:40Z http://hdl.handle.net/2117/15684 Title: Machine learning in human cancer research Authors: Vellido Alcacena, Alfredo; Lisboa, Paulo J.G. Abstract: Evidence-based medicine has grown in stature over three decades and is now regarded a key development of modern medicine. The evidence base can be heterogeneous, involving both qualitative knowledge and measured quantitative data. Data analysis in the area of cancer research has for long been the playing field of statisticians but, over the last decade, Machine Learning (ML) methods have also begun to establish themselves an an alternative and promising approach to computer-based data analysis in oncology. In this chapter, we provide a state-of-the-art in the main areas of cancer research in which ML methods are currently being applied, and discuss some of the advantages and disadvantages of their application. We also comment on and illustrate the integration of ML methods in clinical oncology decision support systems. 2012-03-28T15:19:41Z http://hdl.handle.net/2117/14460 Title: Tracking deformable objects and dealing with same class object occlusion Authors: Alquézar Mancho, René; Amézquita, N; Serratosa Casanelles, Francesc Abstract: This paper presents an extension of a previously reported method for object tracking in video sequences to handle the problems of object crossing and occlusion by other objects in the same class that the one followed. The proposed solution is embedded in a system that integrates recognition and tracking in a probabilistic framework. In a recent work, a method to approach the object occlusion problem was proposed that failed when the object crossed or was occluded by another object of the same class. Here we present an attempt to overcome this limitation and show some promising results. The method is based on the assumption that when two objects cross each other there is not a brusque change of the trajectories. Our system uses object recognition results provided by a neural net that are computed from colour features of image regions for each frame. The location of tracked objects is represented through probability images that are updated dynamically using both recognition and tracking results. From these probabilities and a prediction of the motion of the object in the image, a binary decision is made for each pixel and object. 2012-01-10T19:22:24Z http://hdl.handle.net/2117/13345 Title: Feature and model selection in 1H-MRS single voxel spectra for cancer classification Authors: González Navarro, Félix Fernando; Belanche Muñoz, Luis Antonio Abstract: Machine learning is a powerful paradigm within which to analyze 1HMRS spectral data for the classification of tumour pathologies. An important characteristic of this task is the high dimensionality of the involved data sets. In this work we apply specific feature selection methods in order to reduce the complexity of the problem on two types of 1H-MRS spectral data: long-echo and short-echo time, which present considerable differences in the spectrum for the same cases. The experimental findings show that the feature selection methods enhance the classification performance of the models induced by several off-the-shelf classifiers and are able to offer very attractive solutions both in terms of prediction accuracy and number of involved spectral frequencies. 2011-09-27T10:41:02Z http://hdl.handle.net/2117/13070 Title: Rule-based assistance to brain tumour diagnosis using LR-FIR Authors: Nebot Castells, M. Àngela; Castro Espinoza, Félix Agustín; Vellido Alcacena, Alfredo; Julià Sapé, Margarida; Arús, Carles Abstract: This paper describes a process of rule-extraction from a multi-centre brain tumour database consisting of nuclear magnetic res- onance spectroscopic signals. The expert diagnosis of human brain tumours can benefit from computer-aided assistance, which has to be readily interpretable by clinicians. Interpretation can be achieved through rule extraction, which is here performed using the LR-FIR algorithm, a method based on fuzzy logic. The experimental results of the classification of three groups of tumours indicate in this study that just three spectral frequencies, out of the 195 from a range pre-selected by experts, are enough to represent, in a simple and intuitive manner, most of the knowledge required to discriminate these groups. 2011-07-28T09:34:54Z http://hdl.handle.net/2117/13007 Title: Investigating human cancer with computational intelligence techniques Authors: Vellido Alcacena, Alfredo; Lisboa, Paulo J.G. Abstract: Driven by the growing demand of personalization of medical procedures, data-based, computer-aided cancer research in human patients is advancing at an accelerating pace, providing a broadening landscape of opportunity for Computational Intelligence methods and related techniques. This landscape can be observed from the wide-reaching view of population studies down to the genotype detail. In this introductory chapter, we provide a sweeping glimpse, by no means exhaustive, of the state-of-the-art in this field at the different scales of data measurement and analysis. We do so by focusing mostly on examples from European research, some of which are the matter of the following chapters of the book. 2011-07-19T11:31:09Z http://hdl.handle.net/2117/13006 Title: Clustering and visualization of multivariate time series Authors: Vellido Alcacena, Alfredo; Olier Caparroso, Iván Abstract: The exploratory investigation of multivariate time series (MTS) may become extremely difficult, if not impossible, for high dimensional datasets. Paradoxically, to date, little research has been conducted on the exploration of MTS trough unsupervised clustering and visualization. In this chapter, the authors describe generative topographic mapping through time (GTM-TT), a model with foundations in probability theory that performs such tasks. The standard version of this model has several limitations that limit its applicablility. Here, the authors reformulate it within a Bayesian approach using variational techniques. The resulting variational Bayesian GTM-TT, described in some details, is shown to behave very robustly in the presence of noise in the MTS, helping to avert the poblem of data overfitting. 2011-07-19T11:03:28Z http://hdl.handle.net/2117/10754 Title: Exploratory characterization of a multi-centre 1H-MRS brain tumour database Authors: Vellido Alcacena, Alfredo; Julià Sapé, Margarida; Romero Merino, Enrique; Arús, Carles Abstract: Non-invasive techniques such asMagnetic Resonance Imaging (MRI) and Magnetic Resonance Spectroscopy (MRS) are often required for the diagnosis of tumours for which conclusive biopsies are not commonly available.While radiologists are used to interpretingMRI, many of them are not accustomed to make sense of the biochemical information provided by MRS. In this situation, oncology radiologists may benefit from the use of computer-based support in their decision making. As part of the AIDTumour research project, the analysis of MRS data corresponding to various tumour pathologies is used to assist expert diagnosis. The high dimensionality of the MR spectra might obscure atypical aspects of the data that would jeopardize their automated classification and, as a result, the process of computerbased diagnostic assistance. In this study, we put forward a method to overcome this potential problem that combines methods of visualization through non-linear dimensionality reduction, automatic outlier detection, and radiologists’ expert opinion. 2010-12-27T11:12:28Z