Bayesian Approach for Morphology-Based 2-D Human Motion Capture

This paper presents a novel technique for 2D human motion capture using a single non calibrated camera. The user's five extremities (head, hands and feet) are extracted, labeled and tracked after silhouette segmentation. As they are the minimal number of points that can be used in order to enable whole body gestural interaction, we henceforth refer to these features as crucial points. The crucial point candidates are defined as the local maxima of the geodesic distance with respect to the center of gravity of the actor region that lie on the silhouette boundary. In order to disambiguate the selected crucial points into head, left and right foot and left and right hand classes, we propose a Bayesian framework that combines a MAP approach weighted by a prior model and the intensities of the tracked crucial points. Due to its low computational complexity, the system can run at real-time paces on standard personal computers, with an average error rate range between 2% and 7% in realistic situations, depending on the context and segmentation quality     

SNR scalability based on matching pursuits

In this paper, SNR scalable representations of video signals are studied. The investigated codecs are well suited for communications applications because they are all based on backward motion-compensated predictive coding, which provides the necessary low-delay property. In a very-low bit rate context (VLBR), the matching pursuits (MP) signal representation algorithm is used to represent the displaced frame difference (DFD) of each layer of a multilevel decomposition of the video signal. A number of conventional prediction schemes that can be generalized to any DFD representation technique are considered. They are compared with an original and MP specific DFD prediction method. Two scenari have been considered. In the first scenario, an enhancement layer is built on a base layer that has been encoded using a classical, i.e., nonscalable scheme. In that case, all methods appear to be comparable, In the second scenario, the fact that the base layer is used as a reference for an enhancement layer is taken into account to build it. In that case, the proposed MP prediction method clearly outperforms all other conventional approaches, Additional lessons can be drawn from this work. The same motion vectors can be used in both SNR layers, and the DFD prediction between layers improves coding efficiency. Moreover, the MP representation of the signal enable us to measure the predictability of the high SNR layer DFD from the low SNR layer DFD, i.e., to quantify the part of the low SNR layer information that also belongs to the high SNR layer     

Analog implementation of a Kohonen map with on-chip learning.

Kohonen maps are self-organizing neural networks that classify and quantify n-dimensional data into a one- or two-dimensional array of neurons. Most applications of Kohonen maps use simulations on conventional computers, eventually coupled to hardware accelerators or dedicated neural computers. The small number of different operations involved in the combined learning and classification process, however, makes the Kohonen model particularly suited to a dedicated VLSI implementation, taking full advantage of the parallelism and speed that can be obtained on the chip. A fully analog implementation of a one-dimensional Kohonen map, with on-chip learning and refreshment of on-chip analog synaptic weights, is proposed. The small number of transistors in each cell allows a high degree of parallelism in the operations, which greatly improves the computation speed compared to other implementations. The storage of analog synaptic weights, based on the principle of current copiers, is emphasized. It is shown that this technique can be used successfully for the realization of VLSI Kohonen maps.     

A strategy for optimizing the monodispersity of fusion proteins: application to purification of recombinant HPV E6 oncoprotein

Recombinant production of HPV oncoprotein E6 is notoriouslydifficult. The unfused sequence is produced in inclusion bodies.By contrast, fusions of E6 to the C-terminus of carrier proteinssuch as maltose-binding protein or gluthatione-S-transferaseare produced soluble. However, it has not yet been possibleto purify E6 protein from such fusion constructs. Here, we showthat this was due to the biophysical heterogeneity of the fusionpreparations. We find that soluble MBP-E6 preparations containtwo subpopulations. A major fraction is aggregated and containsexclusively misfolded E6 moieties (`soluble inclusion bodies').A minor fraction is monodisperse and contains the properly foldedE6 moieties. Using monodispersity as a screening criterion,we optimized the expression conditions, the purification processand the sequence of E6, finally obtaining stable monodisperseMBP-E6 preparations. In contrast to aggregated MBP-E6, thesepreparations yielded fully soluble E6 after proteolytic removalof MBP. Once purified, these E6 proteins are stable, foldedand biologically active. The first biophysical measurementson pure E6 were performed. This work shows that solubility isnot a sufficient criterion to check that the passenger proteinin a fusion construct is properly folded and active. By contrast,monodispersity appears as a better quality criterion. The monodispersity-basedstrategy presented here constitutes a general method to preparefusion proteins with optimized folding and biological activity.     

Method for forming two-dimensional paracrystals of biological filaments on lipid monolayers

A method is described for electron microscopic observation of two-dimensional paracrystals on unsupported lipid monolayers. The method uses a hydrophobic holey C-coated grid placed on a monolayer made positively charged by the inclusion of stearylamine (SA) and has been used to align scallop thin filaments and reconstituted actin/tropomyosin filaments to form paracrystals. The use of unsupported monolayers allows the paracrystals to be viewed in either negative stain or with cryoelectron microscopy. Those paracrystals in frozen hydrated specimens have better order than those with negative stain. It was found that varying the lipid composition between the less fluid distearolyphosphotidylcholine/SA and the more fluid egg yolk phosphotidylcholine/SA alters the size and order of the paracrystals, the more fluid system having smaller, more ordered paracrystalline domains. The advantage of the technique for studying actin/thin filaments is the ability to form large two-dimensional paracrystals under physiological conditions of [Mg²⁺] and pH.     

Morphological feature extraction for the classification of digitalimages of cancerous tissues

Presents a new method for automatic recognition of cancerous tissues from an image of a microscopic section. Based on the shape and the size analysis of the observed cells, this method provides the physician with nonsubjective numerical values for four criteria of malignancy. This automatic approach is based on mathematical morphology, and more specifically on the use of geodesy. This technique is used first to remove the background noise from the image and then to operate a segmentation of the nuclei of the cells and an analysis of their shape, their size and their texture. From the values of the extracted criteria, an automatic classification of the image (cancerous or not) is finally operated     

Optimum quality and progressive resolution of video signals

The problem of encoding and transmitting various video formats in a way making possible for all receivers to decode those formats is studied. The quantization for a general subband coder is first optimized using some perceptual criterion. New methods for filtering and decimating are presented, which can be combined with orthogonal transformation. Those operations are analysed. Finally two coding schemes are presented, and simulations are also provided.     

Bibliometric overview of the utilization of artificial neural networks in medicine and biology

The distribution of articles involving artificial neural networks (ANN) in the fields of medicine and biology and appearing in the ISI (Institute for Scientific Information) databases during the period 2000-2001 was analysed. The following parameters were considered: the number of articles, the total impact factor, the ISI journal category, the source country population, and the gross domestic product. Among the 803 articles and the 49 countries considered, the 5 most prolific (in term of the number of publications) were the USA, The United Kingdom, Germany, Italy, and Canada; other active countries included Sweden, Netherlands, Spain, France, Japan, and China. Comparison between the USA and the European Union, and the distribution of ANN publications among the subdisciplines of the life sciences and clinical medicine are also presented.     

Amperometric ATP microbiosensors for the analysis of chemosensitivity at rat carotid bodies

The physiological application of amperometric adenosine triphosphate (ATP) microbiosensors for characterizing the stimulus-response at rat carotid bodies superfused with high potassium concentrations, during normoxic hypercapnia, and during hypoxia is demonstrated using the peripheral arterial chemoreceptors in the carotid body of rats as a model system. Amperometric microbiosensors based on glucose oxidase (GOD) and hexokinase (HEX) immobilized within a polymer matrix at the surface of Pt disk microelectrodes (diameter: 25 microm) are positioned at a distance of approximately 100 microm above the carotid body surface for detecting extracellular ATP. A linear calibration function of ATP microbiosensors in the physiologically relevant concentration range of 0-40 microM ATP enables quantitative detection of ATP released at the carotid body surface in response to physiological stimuli. It is shown that these stimuli induce extracellular ATP release from the carotid body at levels of 4-10 microM. Other electroactive neurotransmitters such as, e.g., catecholamines are coreleased by the carotid body at hypercapnic, hypoxic and high-potassium stimulus, are simultaneously detected utilizing a dual-electrode assembly with an ATP microbiosensor and a second bare channel providing a colocalized reference measurement for ATP quantification.