Optimized combinatorial clustering for stochastic processes

As a new data processing era like Big Data, Cloud Computing, and Internet of Things approaches, the amount of data being collected in databases far exceeds the ability to reduce and analyze these data without the use of automated analysis techniques, data mining. As the importance of data mining has grown, one of the critical issues to emerge is how to scale data mining techniques to larger and complex databases so that it is particularly imperative for computationally intensive data mining tasks such as identifying natural clusters of instances. In this paper, we suggest an optimized combinatorial clustering algorithm for noisy performance which is essential for large data with random sampling. The algorithm outperforms conventional approaches through various numerical and qualitative thresholds like mean and standard deviation of accuracy and computation speed.     

Abstract processes in texture discrimination.

In this study some experiments on texture segmentation are reported using the local Gabor power spectrum. The techniques applied are: (1) supervised pixel classification; (2) boundary detection by spectral dissimilarity estimation; (3) region-based segmentation based on Gaussian spectral estimation; and (4) the same as (3) but based on central moments of the local spectrum. It is shown that very-acceptable-to-excellent results can be obtained. It is argued, however, that the shortcomings of region-based and boundary-based approaches require that both processes should act in parallel, not only in digital image processing but also in the modelling of visual perception.     

Automatic discrimination of fine roots in minirhizotron images

Minirhizotrons provide detailed information on the production, life history and mortality of fine roots. However, manual processing of minirhizotron images is time-consuming, limiting the number and size of experiments that can reasonably be analysed. Previously, an algorithm was developed to automatically detect and measure individual roots in minirhizotron images. Here, species-specific root classifiers were developed to discriminate detected roots from bright background artifacts. Classifiers were developed from training images of peach (Prunus persica), freeman maple (Acer x freemanii) and sweetbay magnolia (Magnolia virginiana) using the Adaboost algorithm. True- and false-positive rates for classifiers were estimated using receiver operating characteristic curves. Classifiers gave true positive rates of 89-94% and false positive rates of 3-7% when applied to nontraining images of the species for which they were developed. The application of a classifier trained on one species to images from another species resulted in little or no reduction in accuracy. These results suggest that a single root classifier can be used to distinguish roots from background objects across multiple minirhizotron experiments. By incorporating root detection and discrimination algorithms into an open-source minirhizotron image analysis application, many analysis tasks that are currently performed by hand can be automated.     

Stimulus class formation and functional equivalence in moderately retarded individuals' conditional discrimination

Three mentally retarded humans first acquired a simple discrimination: Simultaneously displayed visual stimuli A1 and A2 functioned as S+ and S-, respectively. The subjects also acquired a conditional discrimination, learning to select visual stimuli B1 and B2 conditionally upon A1 and A2, respectively. Then, B1 and B2 were displayed without A1 or A2. Subjects selected B1, an emergent discrimination that showed that B1 and B2 had become functionally equivalent to A1 and A2, respectively. Two subjects next learned to select C1 and C2 conditionally upon B1 and B2, respectively. They also learned to select B1 and B2 conditionally upon D1 and D2, respectively. Subsequent simple discrimination probe trials displayed (a) C1 and C2 and (b) D1 and D2. On the former, the subjects nearly always selected C1. On the latter, they initially selected D1 and D2 about equally often. Thus, the emergence of simple discrimination appears to depend on a specific experimental history.     

The resolution and discrimination of images in the visual system

We analysed the conformity of optic and retinal cones anatomy factors by the two-point test. Obtained by F. Campbell and R. W. Gubish, the point spread function has a width of about 1 arc min. Cones sizes are equal to 0.5 arc min in the fovea. Functional pixel consists of 3-5 cones under the point spread function. Such an organisation in very useful in decreasing the samping noise of receptors. We carried out psychophysical investigations to show a consensus among the optic, receptors', and neuronal levels. In experiments we studied changes of the two-point pattern perception in respect to the points separation, measured the orientation threshold of small size stimuli. Data were compared with optical point-spread function, the hexagonal mosaic of cones, and line spread function of spatial elements, which form spatial frequency channels at the cortical level.     

An equivalence result for integral equations with application to branching processes

A theorem is proved, concerning expected values of a multitype branching process in a varying environment. The consequence of the theorem is that the branching process can be treated (in the sense of expected values) as a dynamical system with control terms. This is of importance in situations where the process serves as an abstract model of the dynamics of malignant cells for use in chemotherapy. A simple example of this kind is given. Work supported by the Polish Academy of Sciences Research Problem No. 10.4.3.01.4.     

Learned odor discrimination in Drosophila without combinatorial odor maps in the antennal lobe

A unifying feature of mammalian and insect olfactory systems is that olfactory sensory neurons (OSNs) expressing the same unique odorant-receptor gene converge onto the same glomeruli in the brain [1-7]. Most odorants activate a combination of receptors and thus distinct patterns of glomeruli, forming a proposed combinatorial spatial code that could support discrimination between a large number of odorants [8-11]. OSNs also exhibit odor-evoked responses with complex temporal dynamics [11], but the contribution of this activity to behavioral odor discrimination has received little attention [12]. Here, we investigated the importance of spatial encoding in the relatively simple Drosophila antennal lobe. We show that Drosophila can learn to discriminate between two odorants with one functional class of Or83b-expressing OSNs. Furthermore, these flies encode one odorant from a mixture and cross-adapt to odorants that activate the relevant OSN class, demonstrating that they discriminate odorants by using the same OSNs. Lastly, flies with a single class of Or83b-expressing OSNs recognize a specific odorant across a range of concentration, indicating that they encode odorant identity. Therefore, flies can distinguish odorants without discrete spatial codes in the antennal lobe, implying an important role for odorant-evoked temporal dynamics in behavioral odorant discrimination.     

Hierarchical combinatorial deep learning architecture for pancreas segmentation of medical computed tomography cancer images

Background

Efficient computational recognition and segmentation of target organ from medical images are foundational in diagnosis and treatment, especially about pancreas cancer. In practice, the diversity in appearance of pancreas and organs in abdomen, makes detailed texture information of objects important in segmentation algorithm. According to our observations, however, the structures of previous networks, such as the Richer Feature Convolutional Network (RCF), are too coarse to segment the object (pancreas) accurately, especially the edge.

Method

In this paper, we extend the RCF, proposed to the field of edge detection, for the challenging pancreas segmentation, and put forward a novel pancreas segmentation network. By employing multi-layer up-sampling structure replacing the simple up-sampling operation in all stages, the proposed network fully considers the multi-scale detailed contexture information of object (pancreas) to perform per-pixel segmentation. Additionally, using the CT scans, we supply and train our network, thus get an effective pipeline.

Result

Working with our pipeline with multi-layer up-sampling model, we achieve better performance than RCF in the task of single object (pancreas) segmentation. Besides, combining with multi scale input, we achieve the 76.36% DSC (Dice Similarity Coefficient) value in testing data.

Conclusion

The results of our experiments show that our advanced model works better than previous networks in our dataset. On the other words, it has better ability in catching detailed contexture information. Therefore, our new single object segmentation model has practical meaning in computational automatic diagnosis.

     

Functional equivalence of industrial metal cleaning processes comparison of metal cleaning processes within LCA

In an LCA case study, the three most frequent industrial metal cleaning technologies were assessed: Cleaning based on aqueous cleaning agents, non-halogenated hydrocarbon solvents and halogenated hydrocarbon solvents. Beside optimisation analysis, the comparison of the cleaning processes was a main goal of the study. The function of metal cleaning processes can be described with a set of parameters called functional parameters. In order to compare different cleaning processes within LCA, it is a precondition that all relevant functional parameters be equivalent. However, metal cleaning processes from different companies normally differ in most of the functional parameters and, thus, are not functionally equivalent. Therefore, it is necessary to calculate the material and energy flows of the processes corresponding to a reference function as a basis for comparison. This can be achieved by simulating the processes according to the functional parameters with the help of a process model. For a general comparison of the technologies, it is also necessary to consider the assessed machines having the same level of optimisation and the same scale.     

Microbial pathway engineering for industrial processes: evolution, combinatorial biosynthesis and rational design

Microbial pathway engineering has made significant progress in multiple areas. Many examples of successful pathway engineering for specialty and fine chemicals have been reported in the past two years. Novel carotenoids and polyketides have been synthesized using molecular evolution and combinatorial strategies. In addition, rational design approaches based on metabolic control have been reported to increase metabolic flux to specific products. Experimental and computational tools have been developed to aid in design, reconstruction and analysis of non-native pathways. It is expected that a hybrid of evolutionary, combinatorial and rational design approaches will yield significant advances in the near future.     

Selective stimulus control in discrimination of lateral mirror images by pigeons

Four monocularly and two binocularly viewing pigeons were trained to peck a key when it displayed one stimulus (S+) but not to peck when it displayed another stimulus (S−). S+ and S− were a lateral mirror-image pair of two-coloured stimuli. When tested for transfer with the untrained eye open, two of the monocular birds pecked more during S− than S+, the other two continuing to favour S+. During generalization tests on the wavelength dimension all monocular birds pecked much more often during one S+ colour than during the other. The colour controlling pecking was that displayed on the side of the key facing the open eye during S+ presentations. Both binocular birds developed asymmetrical responses to the key, one favouring the left, the other the right side of the key. Generalization tests on the wavelength dimension showed selective control by the colour displayed on the favoured side of the key during S+ presentations. The results are interpreted as supporting the view that pigeons learn to discriminate lateral mirror images by developing asymmetrical observing responses that convert the left-right difference between the mirror images into a difference more easily discriminable.     

A combinatorial approach to substrate discrimination in the P450 CYP1A subfamily

A comparison of all known mammalian CYP1A sequences identifies nineteen sequence regions that are conserved within all 1A1s or within all 1A2s but at the same time systematically differ between any 1A1 and any 1A2. The purpose of this study was to explore links between these specific CYP1A sequence signatures and substrate specificity shift through the kinetic analysis of combinatorial variants of increasing complexity. The less complex variants correspond to multiple mutations within a short segment of their sequence. The more complex variants correspond to mosaic P450s recombining 1A1 and 1A2 sequences (up to 5 crossovers per sequence). Fifty-eight such functional CYP1A variants and parental wild-type enzymes were expressed in yeast and assayed with 7-alkoxyresorufins and ethoxyflurorescein ethyl ester as substrates. Observed kinetic data were analyzed by multivariate statistical analyses and hierarchical clustering in order to highlight correlations and identify potential sequence-activity relationships within the three-dimensional function space investigated. Several variants are outliers in these representations and show a redistribution of their substrate specificity compared to wild-type CYP1As. Some combinations of sequence elements were identified that significantly discriminate between 1A1 and 1A2 for these three substrates. The comparison of this combinatorial approach with previous results of site-directed mutagenesis is discussed.     

The combinatorial distance geometry method for the calculation of molecular conformation. II. Sample problems and computational statistics

The performance of a branch and bound algorithm for molecular energy minimization is evaluated on a variety of test problems. Although not at present efficient enough for use in most practical situations, we show that it has distinct advantages over more conventional methods of global minimization. In addition, this study illustrates the technique on which the present algorithm is based, and the problems which must be overcome in developing an efficient algorithm based on similar principles.     

Does reduced social contact affect discrimination of distance cues and individual vocalizations?

For songbirds, experience with conspecific vocalizations during development is important for the ontogeny of both production of vocalizations and perceptual abilities in young oscines. We examined the effects of reduced experience with conspecific vocalizations during development on two auditory perceptual tasks: discrimination of distance cues and discrimination between individual vocalizations in black-capped chickadees, Poecile atricapillus. Discrimination of distance cues was nearly identical in field- and isolate-reared chickadees, but the capacity for memorization of numbers of individual vocalizations was lower in isolate-reared chickadees. We conclude that discrimination of distance cues, and thus distance perception, is probably not learned through experience but is an innate skill. We also conclude that discrimination between many individual vocalizations, a task demanding memorization, is aided by early experience with conspecific vocalizations.Copyright 2003 Published by Elsevier Science Ltd on behalf of The Association for the Study of Animal Behaviour.      

Statistical theory of combinatorial libraries of folding proteins: energetic discrimination of a target structure

A self-consistent theory is presented that can be used to estimate the number and composition of sequences satisfying a predetermined set of constraints. The theory is formulated so as to examine the features of sequences having a particular value of Delta=E(f)-(u), where E(f) is the energy of sequences when in a target structure and (u) is an average energy of non-target structures. The theory yields the probabilities w(i)(alpha) that each position i in the sequence is occupied by a particular monomer type alpha. The theory is applied to a simple lattice model of proteins. Excellent agreement is observed between the theory and the results of exact enumerations. The theory provides a quantitative framework for the design and interpretation of combinatorial experiments involving proteins, where a library of amino acid sequences is searched for sequences that fold to a desired structure.     

The combinatorial distance geometry method for the calculation of molecular conformation. I. A new approach to an old problem

A new approach to the long-standing local minimum problem of molecular energy minimization is proposed. The approach relies upon a field of computer mathematics known as combinatorial optimization, together with methods of conformational analysis derived from distance geometry. The advantages over the usual numerical techniques of optimization are, first, that the algorithms derived are globally convergent, and second, that the mathematical problems involved are well-posed and suitable for study within the modern theory of computational complexity. In this paper we introduce the approach, and describe a computer program based on it.     

Parallel and combinatorial approaches for synthesis of ligands

Although new detection screening methods must still be developed, the actual main limitation in combinatorial chemistry seems to be the diversity of ligands that can be generated in terms of real structural and chemical diversity. Thus, there is a strong interest for the development of different strategies for the parallel or combinatorial synthesis of ligands. We report here a selection of recent attempts proposing 'open' approaches able to increase the diversity of molecular architecture truly accessible via parallel or combinatorial processes.