PubMed Health⌕ Search

Biomedical subjects

B De Moor

Publications and source records attributed to B De Moor.

13 recordsLinked to original sources
PubMed2003

Evaluation of the vector space representation in text-based gene clustering.

Thanks to its increasing availability, electronic literature can now be a major source of information when developing complex statistical models where data is scarce or contains much noise. This raises the question of how to deeply integrate information from domain literature with experimental data. Evaluating what kind of statistical text representations can integrate literature knowledge in clustering still remains an unsufficiently explored topic. In this work we discuss how the bag-of-words representation can be used successfully to represent genetic annotation and free-text information coming from different databases. We demonstrate the effect of various weighting schemes and information sources in a functional clustering setup. As a quantitative evaluation, we contrast for different parameter settings the functional groupings obtained from text with those obtained from expert assessments and link each of the results to a biological discussion.

Artificial Intelligence↗
PubMed2003

Probabilistic graphical models for computational biomedicine.

BACKGROUND: As genomics becomes increasingly relevant to medicine, medical informatics and bioinformatics are gradually converging into a larger field that we call computational biomedicine. OBJECTIVES: Developing a computational framework that is common to the different disciplines that compose computational biomedicine will be a major enabler of the further development and integration of this research domain. METHODS: Probabilistic graphical models such as Hidden Markov Models, belief networks, and missing-data models together with computational methods such as dynamic programming, Expectation-Maximization, data-augmentation Gibbs sampling, and the Metropolis-Hastings algorithm provide the tools for an integrated probabilistic approach to computational biomedicine. RESULTS AND CONCLUSIONS: We show how graphical models have already found a broad application in different fields composing computational biomedicine. We also indicate several challenges that lie at the interface between medical informatics, statistical genomics, and bioinformatics. We also argue that graphical models offer a unified framework making it possible to integrate in a statistically meaningful way multiple models ranging from the molecular level to cellular and to clinical levels. Because of their versatility and firm statistical underpinning, we assert that probabilistic graphical models can serve as the lingua franca for many computationally intensive approaches to biology and medicine. As such, graphical models should be a foundation of the curriculum of students in these fields. From such a foundation, students could then build towards specific computational methods in medical informatics, medical image analysis, statistical genetics, or bioinformatics while keeping the communication open between these areas.

Computational Biology↗
PubMed2002

An algorithm including results of gray-scale and power Doppler ultrasound examination to predict endometrial malignancy in women with postmenopausal bleeding.

OBJECTIVE: To determine if power Doppler ultrasound examination of the endometrium can contribute to a correct diagnosis of endometrial malignancy in women with postmenopausal bleeding and endometrium > or = 5 mm. METHODS: Eighty-three women with postmenopausal bleeding and endometrium > or = 5 mm underwent gray-scale and power Doppler ultrasound examination using predetermined, standardized settings. Suspicion of endometrial malignancy at gray-scale ultrasound examination (endometrial morphology) was noted, and the color content of the endometrium at power Doppler examination was estimated subjectively (endometrial color score). Computer analysis of the most vascularized area of the endometrium was done off-line in a standardized manner. Stepwise multivariate logistic regression analysis was carried out to determine which subjective and objective ultrasound and power Doppler variables satisfied the criteria to be included in a model to calculate the probability of endometrial malignancy. RESULTS: Endometrial thickness, vascularity index (vascularized area/endometrial area), and use of hormone replacement therapy (HRT) satisfied the criteria to be included in the model used to calculate the 'objective probability of endometrial malignancy'. Endometrial morphology, endometrial color score and HRT use satisfied the criteria to be included in the model to calculate the 'subjective probability of malignancy'. Endometrial thickness > or = 10.5 mm had a sensitivity with regard to endometrial cancer of 0.88 and a specificity of 0.61. At a fixed sensitivity of 0.88, the specificity of the 'objective probability of malignancy' (0.81) was superior to all other ultrasound and power Doppler variables (P = 0.001-0.02). The 'objective probability of malignancy' detected more malignancies at endometrium 5-15 mm than endometrial morphology (5/7 vs. 1/7, i.e. 0.71 vs. 0.14; P = 0.125) with a similar specificity (49/57 vs. 51/57, i.e. 0.86 vs. 0.89). CONCLUSION: Power Doppler ultrasound can contribute to a correct diagnosis of endometrial malignancy, especially if the endometrium measures 5-15 mm. The use of regression models including power Doppler results to estimate the risk of endometrial cancer deserves further development.

Aged↗
PubMed2002

Bayesian framework for least-squares support vector machine classifiers, gaussian processes, and kernel Fisher discriminant analysis.

The Bayesian evidence framework has been successfully applied to the design of multilayer perceptrons (MLPs) in the work of MacKay. Nevertheless, the training of MLPs suffers from drawbacks like the nonconvex optimization problem and the choice of the number of hidden units. In support vector machines (SVMs) for classification, as introduced by Vapnik, a nonlinear decision boundary is obtained by mapping the input vector first in a nonlinear way to a high-dimensional kernel-induced feature space in which a linear large margin classifier is constructed. Practical expressions are formulated in the dual space in terms of the related kernel function, and the solution follows from a (convex) quadratic programming (QP) problem. In least-squares SVMs (LS-SVMs), the SVM problem formulation is modified by introducing a least-squares cost function and equality instead of inequality constraints, and the solution follows from a linear system in the dual space. Implicitly, the least-squares formulation corresponds to a regression formulation and is also related to kernel Fisher discriminant analysis. The least-squares regression formulation has advantages for deriving analytic expressions in a Bayesian evidence framework, in contrast to the classification formulations used, for example, in gaussian processes (GPs). The LS-SVM formulation has clear primal-dual interpretations, and without the bias term, one explicitly constructs a model that yields the same expressions as have been obtained with GPs for regression. In this article, the Bayesian evidence framework is combined with the LS-SVM classifier formulation. Starting from the feature space formulation, analytic expressions are obtained in the dual space on the different levels of Bayesian inference, while posterior class probabilities are obtained by marginalizing over the model parameters. Empirical results obtained on 10 public domain data sets show that the LS-SVM classifier designed within the Bayesian evidence framework consistently yields good generalization performances.

Artificial Intelligence↗
PubMed2001

Asymptotic relative entropy of entanglement.

We present an analytical formula for the asymptotic relative entropy of entanglement with respect to positive partial transpose states for Werner states of arbitrary dimension. We then demonstrate its validity using methods from convex optimization. This is the first case in which the asymptotic value of a subadditive entanglement measure has been calculated.

Journal Article↗
PubMed2001

Optimal control by least squares support vector machines.

Support vector machines have been very successful in pattern recognition and function estimation problems. In this paper we introduce the use of least squares support vector machines (LS-SVM's) for the optimal control of nonlinear systems. Linear and neural full static state feedback controllers are considered. The problem is formulated in such a way that it incorporates the N-stage optimal control problem as well as a least squares support vector machine approach for mapping the state space into the action space. The solution is characterized by a set of nonlinear equations. An alternative formulation as a constrained nonlinear optimization problem in less unknowns is given, together with a method for imposing local stability in the LS-SVM control scheme. The results are discussed for support vector machines with radial basis function kernel. Advantages of LS-SVM control are that no number of hidden units has to be determined for the controller and that no centers have to be specified for the Gaussian kernels when applying Mercer's condition. The curse of dimensionality is avoided in comparison with defining a regular grid for the centers in classical radial basis function networks. This is at the expense of taking the trajectory of state variables as additional unknowns in the optimization problem, while classical neural network approaches typically lead to parametric optimization problems. In the SVM methodology the number of unknowns equals the number of training data, while in the primal space the number of unknowns can be infinite dimensional. The method is illustrated both on stabilization and tracking problems including examples on swinging up an inverted pendulum with local stabilization at the endpoint and a tracking problem for a ball and beam system.

Feedback↗
PubMed2001

A higher-order background model improves the detection of promoter regulatory elements by Gibbs sampling.

MOTIVATION: Transcriptome analysis allows detection and clustering of genes that are coexpressed under various biological circumstances. Under the assumption that coregulated genes share cis-acting regulatory elements, it is important to investigate the upstream sequences controlling the transcription of these genes. To improve the robustness of the Gibbs sampling algorithm to noisy data sets we propose an extension of this algorithm for motif finding with a higher-order background model. RESULTS: Simulated data and real biological data sets with well-described regulatory elements are used to test the influence of the different background models on the performance of the motif detection algorithm. We show that the use of a higher-order model considerably enhances the performance of our motif finding algorithm in the presence of noisy data. For Arabidopsis thaliana, a reliable background model based on a set of carefully selected intergenic sequences was constructed. AVAILABILITY: Our implementation of the Gibbs sampler called the Motif Sampler can be used through a web interface: http://www.esat.kuleuven.ac.be/~thijs/Work/MotifSampler.html. CONTACT: gert.thijs@esat.kuleuven.ac.be; yves.moreau@esat.kuleuven.ac.be

Algorithms↗
PubMed2001

Improved long-term temperature prediction by chaining of neural networks.

When an artificial neural network (ANN) is trained to predict signals p steps ahead, the quality of the prediction typically decreases for large values of p. In this paper, we compare two methods for prediction with ANNs: the classical recursion of one-step ahead predictors and a new kind of chain structure. When applying both techniques to the prediction of the temperature at the end of a blast furnace, we conclude that the chaining approach leads to an improved prediction of the temperature and avoidance of instabilities, since the chained networks gradually take the prediction of their predecessors in the chain as an extra input. It is observed that instabilities might occur in the iterative case, which does not happen with the chaining approach. To select relevant inputs and decrease the number of weights in this approach, Automatic Relevance Determination (ARD) for multilayer perceptrons is applied.

Bayes Theorem↗
PubMed2000

Fetal electrocardiogram extraction by blind source subspace separation.

In this paper, we propose the emerging technique of independent component analysis, also known as blind source separation, as an interesting tool for the extraction of the antepartum fetal electrocardiogram from multilead cutaneous potential recordings. The technique is illustrated by means of a real-life example.

Algorithms↗
PubMed1999

Artificial neural network models for the preoperative discrimination between malignant and benign adnexal masses.

OBJECTIVE: The aim of this study was to generate and evaluate artificial neural network (ANN) models from simple clinical and ultrasound-derived criteria to predict whether or not an adnexal mass will have histological evidence of malignancy. DESIGN: The data were collected prospectively from 173 consecutive patients who were scheduled to undergo surgical investigations at the University Hospitals, Leuven, between August 1994 and August 1996. The outcome measure was the histological classification of excised tissues as malignant (including borderline) or benign. METHODS: Age, menopausal status and serum CA 125 levels and sonographic features of the adnexal mass were encoded as variables. The ANNs were trained on a randomly selected set of 116 patient records and tested on the remainder (n = 57). The performance of each model was evaluated using receiver operating characteristic (ROC) curves and compared with corresponding data from an established risk of malignancy index (RMI) and a logistic regression model. RESULTS: There were 124 benign masses, five of borderline malignancy and 44 invasive cancers (of which 29% were metastatic); 37% of patients with a malignant or borderline tumor had stage I disease. The best ANN gave an area under the ROC curve of 0.979 for the whole dataset, a sensitivity of 95.9% and specificity of 93.5%. The corresponding values for the RMI were 0.882, 67.3% and 91.1%, and for the logistic regression model 0.956, 95.9% and 85.5%, respectively. CONCLUSION: An ANN can be trained to provide clinically accurate information, on whether or not an adnexal mass is malignant, from the patient's menopausal status, serum CA 125 levels, and some simple ultrasonographic criteria.

Adnexal Diseases↗
PubMed1994

Subclavian vein hemodialysis catheters: advantages and disadvantages.

The pros and cons of subclavian vein hemodialysis catheters are reviewed. The subclavian vein catheter offers rapid and adequate vascular access. Other advantages are preservation of patient mobility and the ease with which the dressing can be secured. In contrast, subclavian vein stenosis has recently been identified as an important long-term complication of subclavian vein catheters that seriously compromises the creation and long-term viability of later ipsilateral arteriovenous fistulas. For this reason, we recommend restricting use of the subclavian hemodialysis catheter for acute situations and preferential use of internal jugular catheters for more chronic approaches because they better preserve venous integrity.

Catheterization, Central Venous↗
PubMed1990

Comparison of SVD methods to extract the foetal electrocardiogram from cutaneous electrode signals.

The paper presents and compares three methods making use of the singular value decomposition (SVD) of a matrix to extract the foetal electrocardiogram (FECG) from cutaneously recorded electrode signals. The first method constructs a set of orthogonal foetal signals (the so-called principal foetal signals) from the recordings, but needs electrode positions far from the foetal heart, in addition to the abdominal electrodes that pick up a mixture of maternal and foetal electrocardiogram. An online adaptive algorithm has been developed such that a real-time implementation becomes feasible. The second method is a new online approach to a technique presented by van Oosterom. Although this method has some important drawbacks and is suboptimal as far as foetal signal-to-noise ratio is concerned, it is still very useful when only a foetal trigger is required, as the signal obtained is not a complete FECG. Finally, a third method is proposed, based on the generalised SVD and interpreted with the new concept of oriented signal-to-signal ratio. An online version is also presented for this method and some results are shown.

Electrocardiography↗