Entwicklung des Pauling-Verfahrens zur Regeneration von Abfallschwefelsäure

Development of the Pauling process for regeneration of spent sulphuric acid . The Pauling rectification process has proved a successful method of regenerating spent sulphuric acid. The further development of the process in the last ten years has included the production of cast iron materials with improved corrosion resistance for use in the boiling vessels. Other development work concerns the operation of the stirrers and the apparatus, the construction of stripping columns, and the introduction of new types of cooling equipment for the hot concentrated sulphuric acid. Furthermore, the scrubbing of the exhaust gases leaving the Pauling apparatus must be adapted to more stringent demands, also with regard to the increasingly varied composition of the spent acids.     

Downregulation of eRF1 by RNA interference increases mis-acylated tRNA suppression efficiency in human cells

The site-specific incorporation of non-natural amino acids into proteins by nonsense suppression has been widely used to investigate protein structure and function. Usually this technique exhibits low incorporation efficiencies of non-natural amino acids into proteins. We describe for the first time an approach for achieving an increased level of nonsense codon suppression with synthetic suppressor tRNAs in cultured human cells. We find that the intracellular concentration of the eukaryotic release factor 1 (eRF1) is a critical parameter influencing the efficiency of amino acid incorporation by nonsense suppression. Using RNA interference we were able to lower eRF1 gene expression significantly. We achieved a five times higher level of amino acid incorporation as compared with non-treated control cells, as demonstrated by enhanced green fluorescent protein (EGFP) fluorescence recovery after importing a mutated reporter mRNA together with an artificial amber suppressor tRNA.     

Occlusion as a measure for visual workload: an overview of TNO occlusion research in car driving

A process-oriented approach by systematically studying driver performance, distraction, and workload is the way to go for assessing safety effects of new telematics applications in vehicles. These systems may strive for drivers' attention and possibly lead to distraction from the primary task. Visual occlusion techniques appear to be an effective means of studying drivers' visual information processing performance. Studies to illustrate this include identifying the minimum visual information drivers need for driving (asking for visual information) and the evaluation of driver support systems such as heading control and adaptive cruise control with respect to visual workload. In other applications the occlusion technique is used to exclude 'visual array' information from the optic flow field. The temporary exclusion of part of the visual field of view was applied in evaluating the effects of different types of driver's side rearview mirrors.     

Volumetric analysis of mice lungs in a clinical magnetic resonance imaging scanner

Small animal models are widely used to study various pathologies. Magnetic resonance imaging (MRI) allows investigation of these animals in a non-invasive way. Therefore, the aim of our study was to develop and evaluate a low-cost approach to measure lung volumes in small animal MRI using a clinical scanner and a specially designed RF coil. Five mice (three of an established emphysema model and two controls) were investigated in a 1.0-T clinical scanner using a specially built small animal saddle coil and three different three-dimensional sequences; overall imaging time was approximately 16 min. Lung volumes were calculated from these images using an interactive watershed transform algorithm for semi-automatic image segmentation. The gold standard for the volume measurement was water displacement after surgical explantation. MRI measured volumes correlated significantly with ex vivo measurements on the explanted lungs (r = 0.99 to 0.89; p < 0.05). Mean lung volume in emphysema model mice was larger than in controls. High-resolution, small animal MRI using a clinical scanner is feasible for volumetric analysis and provides an alternative to a dedicated small animal scanner.     

Photoreceptor proteins, "star actors of modern times": a review of the functional dynamics in the structure of representative members of six different photoreceptor families

Six well-characterized photoreceptor families function in Nature to mediate light-induced signal transduction: the rhodopsins, phytochromes, xanthopsins, cryptochromes, phototropins, and BLUF proteins. The first three catalyze E/Z isomerization of retinal, phytochromobilin, and p-coumaric acid, respectively, while the last three all have a different flavin-based photochemistry. For many of these photoreceptor proteins, (many of) the details of the conversion of the light-induced change in configuration of their chromophore into a signaling state and eventually a biological response have been resolved. Some members of the rhodopsins, the xanthopsins, and the phototropins are so well characterized that they function as model systems to study (receptor) protein dynamics and (un)folding.     

Ensembles of classifiers for handwritten word recognition

Handwritten text recognition is one of the most difficult problems in the field of pattern recognition. Recently, a number of classifier creation and combination methods, known as ensemble methods, have been proposed in the field of machine learning. They have shown improved recognition performance over single classifiers. In this paper the application of some of those ensemble methods in the domain of offline cursive handwritten word recognition is described. The basic word recognizers are given by hidden Markov models (HMMs). It is demonstrated through experiments that ensemble methods have the potential of improving recognition accuracy also in the domain of handwriting recognition.Received: 23 November 2001, Accepted: 19 September 2002, Published online: 6 June 2003     

Automatic learning of cost functions for graph edit distance

Graph matching and graph edit distance have become important tools in structural pattern recognition. The graph edit distance concept allows us to measure the structural similarity of attributed graphs in an error-tolerant way. The key idea is to model graph variations by structural distortion operations. As one of its main constraints, however, the edit distance requires the adequate definition of edit cost functions, which eventually determine which graphs are considered similar. In the past, these cost functions were usually defined in a manual fashion, which is highly prone to errors. The present paper proposes a method to automatically learn cost functions from a labeled sample set of graphs. To this end, we formulate the graph edit process in a stochastic context and perform a maximum likelihood parameter estimation of the distribution of edit operations. The underlying distortion model is learned using an Expectation Maximization algorithm. From this model we finally derive the desired cost functions. In a series of experiments we demonstrate the learning effect of the proposed method and provide a performance comparison to other models.