Bayesian feature and model selection for Gaussian mixture models

We present a Bayesian method for mixture model training that simultaneously treats the feature selection and the model selection problem. The method is based on the integration of a mixture model formulation that takes into account the saliency of the features and a Bayesian approach to mixture learning that can be used to estimate the number of mixture components. The proposed learning algorithm follows the variational framework and can simultaneously optimize over the number of components, the saliency of the features, and the parameters of the mixture model. Experimental results using high-dimensional artificial and real data illustrate the effectiveness of the method.     

The global k-means clustering algorithm

We present the global k-means algorithm which is an incremental approach to clustering that dynamically adds one cluster center at a time through a deterministic global search procedure consisting of N (with N being the size of the data set) executions of the k-means algorithm from suitable initial positions. We also propose modifications of the method to reduce the computational load without significantly affecting solution quality. The proposed clustering methods are tested on well-known data sets and they compare favorably to the k-means algorithm with random restarts.     

A divide-and-conquer method for multi-net classifiers

Several researchers have shown that substantial improvements can be achieved in difficult pattern recognition problems by combining the outputs of multiple neural networks. In this work, we present and test a pattern classification multi-net system based on both supervised and unsupervised learning. Following the ‘divide-and-conquer’ framework, the input space is partitioned into overlapping subspaces and neural networks are subsequently used to solve the respective classification subtasks. Finally, the outputs of individual classifiers are appropriately combined to obtain the final classification decision. Two clustering methods have been applied for input space partitioning and two schemes have been considered for combining the outputs of the multiple classifiers. Experiments on well-known data sets indicate that the multi-net classification system exhibits promising performance compared with the case of single network training, both in terms of error rates and in terms of training speed (especially if the training of the classifiers is done in parallel). ID="A1"Correspondance and offprint requests to: D. Frosyniotis, National Technical University of Athens, Department of Electrical and Computer Engineering, Zographou 157 73, Athens, Greece. E-mail: andreas@cs.ntua.gr     

Theoretical and experimental investigation of pulsed laser grooving process

A theoretical model has been developed for simulating the laser grooving process. It takes into account the interaction among subsequent pulses, the required time for the melting temperature to be reached and the subsequent removal of a finite volume of material during each laser pulse. The model predicts the maximum groove depth that can be achieved for a specified set of process parameters, such as laser power, pulsing frequency, and scanning velocity. The theoretical predictions have been experimentally tested with a medium-power laser beam.     

A new method for the control of the river Nile

This paper introduces a new stochastic control method, Extended Linear Quadratic Gaussian control, that appears to be particularly promising for the management of complex reservoir configurations such as the River Nile system. The method is introduced by means of a short case study related to the regulation of the Equatorial Lakes in the Nile basin. If the necessary data are available, the method can be easily adapted to the entire Nile system and used to investigate a variety of management issues.     

Fibro/chondrogenic differentiation of dental stem cells into chitosan/alginate scaffolds towards temporomandibular joint disc regeneration

Tissue engineering (TE) may provide effective alternative treatment for challenging temporomandibular joint (TMJ) pathologies associated with disc malpositioning or degeneration and leading to severe masticatory dysfunction. Aim of this study was to evaluate the potential of chitosan/alginate (Ch/Alg) scaffolds to promote fibro/chondrogenic differentiation of dental pulp stem cells (DPSCs) and production of fibrocartilage tissue, serving as a replacement of the natural TMJ disc. Ch/Alg scaffolds were fabricated by crosslinking with CaCl2 combined or not with glutaraldehyde, resulting in two scaffold types that were physicochemically characterized, seeded with DPSCs or human nucleus pulposus cells (hNPCs) used as control and evaluated for cell attachment, viability, and proliferation. The DPSCs/scaffold constructs were incubated for up to 8 weeks and assessed for extracellular matrix production by means of histology, immunofluorescence, and thermomechanical analysis. Both Ch/Alg scaffold types with a mass ratio of 1:1 presented a gel-like structure with interconnected pores. Scaffolds supported cell adhesion and long-term viability/proliferation of DPSCs and hNPCs. DPSCs cultured into Ch/Alg scaffolds demonstrated a significant increase of gene expression of fibrocartilaginous markers (COLI, COL X, SOX9, COM, ACAN) after up to 3 weeks in culture. Dynamic thermomechanical analysis revealed that scaffolds loaded with DPSCs significantly increased storage modulus and elastic response compared to cell-free scaffolds, obtaining values similar to those of native TMJ disc. Histological data and immunochemical staining for aggrecan after 4 to 8 weeks indicated that the scaffolds support abundant fibrocartilaginous tissue formation, thus providing a promising strategy for TMJ disc TE-based replacement.

Open image in new window

     

A kurtosis-based dynamic approach to Gaussian mixture modeling

We address the problem of probability density function estimation using a Gaussian mixture model updated with the expectation-maximization (EM) algorithm. To deal with the case of an unknown number of mixing kernels, we define a new measure for Gaussian mixtures, called total kurtosis, which is based on the weighted sample kurtoses of the kernels. This measure provides an indication of how well the Gaussian mixture fits the data. Then we propose a new dynamic algorithm for Gaussian mixture density estimation which monitors the total kurtosis at each step of the EM algorithm in order to decide dynamically on the correct number of kernels and possibly escape from local maxima. We show the potential of our technique in approximating unknown densities through a series of examples with several density estimation problems     

Visual tracking using the Earth Mover's Distance between Gaussian mixtures and Kalman filtering

In this paper, we demonstrate how the differential Earth Mover's Distance (EMD) may be used for visual tracking in synergy with Gaussian mixtures models (GMM). According to our model, motion between adjacent frames results in variations of the mixing proportions of the Gaussian components representing the object to be tracked. These variations are computed in closed form by minimizing the differential EMD between Gaussian mixtures, yielding a very fast algorithm with high accuracy, without recurring to the EM algorithm in each frame. Moreover, we also propose a framework to handle occlusions, where the prediction for the object's location is forwarded to an adaptive Kalman filter whose parameters are estimated on line by the motion model already observed. Experimental results show significant improvement in tracking performance in the presence of occlusion.     

Neural-network methods for boundary value problems with irregularboundaries

Partial differential equations (PDEs) with boundary conditions (Dirichlet or Neumann) defined on boundaries with simple geometry have been successfully treated using sigmoidal multilayer perceptrons in previous works. The article deals with the case of complex boundary geometry, where the boundary is determined by a number of points that belong to it and are closely located, so as to offer a reasonable representation. Two networks are employed: a multilayer perceptron and a radial basis function network. The later is used to account for the exact satisfaction of the boundary conditions. The method has been successfully tested on two-dimensional and three-dimensional PDEs and has yielded accurate results.     

Surface tension study of the interaction between two nonionic surfactants and two direct dyes

The complex-forming interaction between two direct dyes, CI Yellow 106 and CI Blue 78, and two nonionic surfactants, namely, polyoxyethylene nonyl phenyl ether (NP-12) and polyoxyethylene stearylamine (R-11), in aqueous solutions was studied. The estimation was made by measuring surface tension and critical micelle concentration (CMC) changes as a function of dye concentration. A decrease of NP-12 surface tension at low concentration and an increase of R-11 surface tension in the presence of both dyes were observed along with a significant decrease of NP-12 CMC values. A difference in the spectrophotometric absorbance of dye solutions in the presence of both surfactants, indicating a change in the environment of the dye chromophore, was also visible. These results confirm the formation of hydrophobic complexes of NP-12 and hydrophilic complexes of R-11 with both dyes.