Jacky Desachy’s research while affiliated with University Hospital Center Pointe-à-Pitre and other places

In this paper, we present an application of the belief function theory for the classification of physiological states in a bioprocess. It also takes account of the relevance of the data sources. The notion of conflict is used to evaluate the relevance of each data source. Another measure of conflict, based on a distance, is also used, and provides globally, better results than the classical notion of conflict used in the Dempster rule of fusion. Experimental results are presented for a bioprocess and show that, with the use of relevance, the results of classification are better.

Today, the pace of progress in fermentation is fast and furious, particularly since the advent of genetic engineering and the recent advances in computer sciences and process control. The high cost associated with many fermentation processes makes optimization of bioreactor performance trough command control very desirable. Clearly, control of fermentation is recognized as a vital component in the operation and successful production of many industries. Today’s advances in measurement, data acquisition and handling technologies provide a wealth of new data which can be used to improve existing models. In this article we propose a method of physiological state identification based on segmentation of bioreactor sensors signals. The underlying of this method is based on the detection of signals singularities by the Maximum of Modulus of Wavelets Transform and their characterization by Hölder exponent evaluation. The physiological states identification is based on the correlation product between biochemical signals. The efficiency of the method has been tested in a fed-batch fermentation having the goal to increase the biomass production.

In this paper we use Markov Models to classify automatically spoligotypes. A spoligotype is a sequence of 43 binary values provided by a DNA analysis technique. These methods, robust and well adapted to sequential data, allow us to generate a model on the basis of probabilities, calculated directly on the observations. We use these techniques to create one classifier for each searched class.

In this paper, we present an application of the evidence theory for the classification of physiological states in a bioprocess. We are particularly interested by the relevance of the data sources which are here biochemical parameters measured during the bioprocess. The evidence theory, and more particularly the notion of conflict is used to evaluate the relevance of each data source. An other measure of conflict, based on a distance, is also used, and provides in some cases, better results than the classical notion of conflict of the evidence theory. Results are presented for two kinds of bioprocesses : batch process (which corresponds to a supervised classification) and fed-batch process (which corresponds to an unsupervised classification). Dans cet article, nous présentons une application de la théorie des fonctions de croyance pour la classification d’états physiologiques dans un bioprocédé. Nous nous intéressons surtout à la pertinence des sources d’informations qui sont ici des paramètres biochimiques mesurés durant le procédé. La théorie des fonctions de croyance, et plus particulièrement la notion de conflit est utilisée pour évaluer la pertinence de chaque source d’information. Une autre mesure du conflit, basée sur une distance, est utilisée comme alternative, et fournit dans certains cas, des résultats plus cohérents qu’avec le conflit défini dans la théorie de Demspter et Shafer. Les résultats concernant deux types de bioprocédés (procédé batch correspondant à une classification supervisée, et procédé fed-batch correspondant à une classification non supervisée) sont présentés.

A new graph matching approach based on 1D information is presented. Each node of the matched graphs represents a fuzzy region (fuzzy segmentation step). Each couple of nodes is linked by a relational histogram which can be assumed to the attraction of two regions following a set of directions. This attraction is computed by a continuous function, depending on the distance of the matched objects. Each case of the histogram corresponds to a particular direction. Then, relational graph computed from strong scenes are matched.

Image matching is a stage one performs as soon as one has two images of the same scene, taken from two different points of view. Matching these images aims at finding the mathematical transformation that enables passing from any point of the first one to the corresponding point in the other. As this study is related to satellite images, we show that the geometrical transformation can be approximated by a homography. Furthermore we want to match two clusters of points with no information of radiometry. Therefore, we have to guess the right parameters for this homography, by minimizing an appropriate cost function we define here. Then, the topography of the cost function is our main concern for the minimisation process. If looking for the right mathematical parameters seems the most natural way, we show that in this case the cost function has “chaotic” variations, so we need a complex technique for the minimization. To avoid this, we suggest guessing the parameters determining the conditions of the snapshot. Thus, we give the expression of the homography from these “physical parameters” and show that the topography of the cost function gets smoother. Thus the minimization process gets simpler.

Fermentation is a critical process for the production simpler substances from organic molecules or could be used to obtain the ethanol by the anaerobic breaking down of sugar. In this work we present different image analysis steps in order to characterize the cell morphology during the biotechnology process. The cell morphology is an important element of the stress which could disturb the production of the biomass or a metabolite. For this purpose we develop a Java software dedicated to an automatic analysis. The software allows us to have information about the growth cells, morphometric analysis (volume/surface) and morphology (budding cells). We want to have the information in biological real time to be able to modify the control parameters, therefore the image is cut into small slices which are analyzed separately by a parallel algorithm. Full Text at Springer, may require registration or fee

For analysis and modelling of the biotechnological process we must look deeper to the biological systems. Everybody knows that cell metabolism and the resulting kinetic is a complex process which could not be modelling completely by a non-linear differential system. The goal of all modelling is either the biocontrol or finding the physiological states. We want to detect the physiological states using a small number of measured signals. We present in this paper the analyses of biochemical parameters using the evidence theory. The evidence theory is also used to characterize the pertinence of the parameters. This pertinence is based on the notion of conflict. We show that our measure of conflict based on a distance provides more coherent results as the classical methods. Based on the analysis of microbiological process a method has been implemented for fermentation real-time analysis.

This paper presents a methodology to detect online the physiological state of strains in a bioreactor. The biologic reactions means that one of the pathways of the metabolism is activated and in this case the microorganisms will produce or will consummate the discrete-event system (DES) is synthesized applying the maximum of modulus of the wavelet transform on measured signals constrained to the biotechnologist expert validation. The determination of holder coefficient by differential evolutionary algorithms allows to make the difference between different discontinuities and to obtain the segmentation of the signals. All these evaluations lead to associate the signals variations during the time to physiological states.

In this paper, we propose several methods to update and upgrade GIS using remote sensing images. In the first part, we present a matching method of GIS vectors on SPOT images, allowing the localisation of geographic linear elements to be improved; and in the second part, we present an extraction method of new linear elements using Ikonos images, with a multi-resolution approach.