Second order cone programming approaches for handling missing and uncertain data

We propose a novel second order cone programming formulation for designing robust classifiers which can handle uncertainty in observations. Similar formulations are also derived for designing regression functions which are robust to uncertainties in the regression setting. The proposed formulations are independent of the underlying distribution, requiring only the existence of second order moments. These formulations are then specialized to the case of missing values in observations for both classification and regression problems. Experiments show that the proposed formulations outperform imputation.

Macrophages in regressed and progressed uveal melanoma

Uveal melanoma (UM) is the most common primary ocular malignancy in adults. In Finland, approximately 50 new cases are diagnosed yearly. Up to 50% of UM metastasize, mostly to the liver, although other organs are also affected. Despite improvements in the management of the primary tumour, the survival rates of patients with metastatic UM are poor. Until the 1970s, UMs were treated by enucleation i.e. removal of the eye. Currently, UM is usually treated by brachytherapy, which is known to influence tumour cells and blood vessels. UMs enucleated both primarily and secondarily after brachytherapy contain tumour-infiltrating macrophages, and a high number of macrophages in primary UM is associated with a shorter survival and a higher microvascular density (MVD) within the tumour tissue. The latter is independently associated with a shorter time to metastatic death. Macrophages have several diverse roles depending on their response to variable signals from the surrounding microenvironment. They function as scavengers, as producers of angiogenic and growth factors as well as proteases, which modulate extracellular matrix. Thus, tumour invasiveness and the risk for metastasis increase with increasing macrophage density. The aim of this study was to evaluate the effects of regression and progression of UM on macrophage numbers and microcirculation factors. Tumour regression is induced by primary brachytherapy, and tumour progression is evidenced by the development of metastases. Understanding the biological behaviour of UMs in the both states may help us in finding new treatment modalities against this disease. To achieve these aims case-control analyses of irradiated UMs and primarily-enucleated eyes (34 matched pairs) were performed. UMs were stained immunohistochemically to detect macrophages, extravascular matrix (EVM) loops and networks, and MVD. Following brachytherapy, a lower MVD was observed. The average number of macrophages remained unchanged. Considering that irradiated melanomas may still contain proliferating tumour cells, a clinically-relevant consequence of my study would be the reassurance that the risk for metastasis is likely to be reduced, given that the low MVD in untreated UMs indicates a favourable prognosis. The effect of progression on macrophages was studied in a paired analysis of primarily-enucleated UM and their corresponding hepatic metastases (48 pairs). A cross-sectional histopathological analysis of these pairs was carried out by staining both specimens in a similar way to the first study. MVD was greater in hepatic metastases than in corresponding primary tumours, and the survival of the patient tended to be shorter if hepatic metastases had a higher MVD. Hepatic metastases had also more dendritic macrophages than the primary UMs. Thus, the progression to metastasis seems to alter the inflammatory status within the tumour. Furthermore, determining MVD of biopsied hepatic metastases may serve as a supplementary tool in estimating the prognosis of patients with metastatic uveal melanoma. After irradiation, the majority of treated eyes have been clinically observed to have pigmented episcleral deposits. A noncomparative clinical case series of 211 irradiated UM eyes were studied by recording the number and location of pigmented episcleral deposits during follow-up visits after brachytherapy. For the first time, the study described pigmented episcleral deposits, which are found in the most UM eyes after brachytherapy, and proved them to consist of macrophages full with engulfed melanin particles. This knowledge may save patients from unnecessary enucleation, because episcleral pigmented deposits might be mistaken for extrascleral tumour growth. The presence of pigmented macrophage-related episcleral deposits was associated with plaque size and isotope rather than with tumour size, suggesting that, in addition to tumour regression, radiation atrophy of retinal pigment epithelium and choroid contributes to the formation of the deposits. In the paired (the same 34 pairs as in the first study) cross-sectional study of irradiated and non-irradiated UMs, clinically-visible episcleral deposits and migrating macrophages in other extratumoral tissues were studied histopathologically. Resident macrophages were present in extratumoral tissues in eyes with both irradiated and non-irradiated UM. Irradiation increased both the number of CD68+ macrophages in the sclera beneath the tumour and the number of clinically-observed episcleral macrophages aggregates. Brachytherapy seemed to alter the route of migration of macrophages: after irradiation, macrophages migrated preferentially through the sclera while in non-irradiated UMs they seemed to migrate more along the choroid. In order to understand the influence of these routes on tumour progression and regression in the future, labelling and tracking of activated macrophages in vivo is required.

Semi-Supervised Classification Using Sparse Gaussian Process Regression

Gaussian Processes (GPs) are promising Bayesian methods for classification and regression problems. They have also been used for semi-supervised learning tasks. In this paper, we propose a new algorithm for solving semi-supervised binary classification problem using sparse GP regression (GPR) models. It is closely related to semi-supervised learning based on support vector regression (SVR) and maximum margin clustering. The proposed algorithm is simple and easy to implement. It gives a sparse solution directly unlike the SVR based algorithm. Also, the hyperparameters are estimated easily without resorting to expensive cross-validation technique. Use of sparse GPR model helps in making the proposed algorithm scalable. Preliminary results on synthetic and real-world data sets demonstrate the efficacy of the new algorithm.

Microsite occupancy and spatial structure of regeneration in three late-successional Norway spruce forests in northern Europe

A comparison of microsite occupancy and the spatial structure of regeneration in three areas of late-successional Norway spruce dominated forest. Pallas-Ylläs is understood to have been influenced only by small-scale disturbance; Dvina-Pinega has had sporadic larger-scale disturbances; Kazkim has been affected by fire. All spruce and birch trees with diameter at breast height (DBH) ?10 cm were mapped in five stands on 40 m x 400 m transects, and those with DBH < 10 cm on 2 or 4 m x 400 m subplots. Microsite type was inventoried at 1m intervals along the centre line and for each tree with DBH < 10 cm. At all study areas small seedlings (h < 0.3 m, DBH < 10 cm) preferentially occupied disturbed microsites. In contrast, spruce saplings (h ? 1.3 m, DBH <10 cm) at all study areas showed less, or no, preference. At Pallas-Ylläs spruce seedlings (h < 1.3 m, DBH < 10 cm) and saplings (h ? 1.3 m, DBH < 10 cm) exhibited spatial correlation at scales from 32-52 m. At Dvina-Pinega saplings of both spruce and birch exhibited spatial correlation at scales from 32-81 m. At Kazkim spatial correlation of seedlings and saplings of both species was exhibited over variable distances. No spatial cross-correlation was found between overstorey basal area (DBH ? 10 cm) and regeneration (h ? 1.3 m, DBH < 10 cm) at any study area. The results confirm the importance of disturbed microsites for seedling establishment, but suggest that undisturbed microsites may sometimes be more advantageous for long-term tree survival. The regeneration gap concept may not be useful in describing the regeneration dynamics of late-successional boreal forests.

The status and use of tree biomass in a semi-arid village ecosystem

The status of the tree biomass resource was investigated in Ungra, a semi-arid village ecosystem in South India. There were 57 tree species with 12 trees capita−1 and 35 trees ha−1. Multiple benefit yielding local tree species dominated the village ecosystem, while fuel only or single end use trees accounted for a small proportion of trees. The standing tree biomass is adequate to meet the requirement of biomass fuels for cooking only for about two years. Village tree biomass is presently being depleted largely for export to urban areas. Tree regeneration is now characterized by transformation from multiple-use local tree species to a few single-use species. A large potential exists for tree biomass production along field boundaries (bunds), stream banks and roadsides. Biomass estimation equations were developed for 10 species.

On the basin-scale detection and attribution of human-induced climate change in monsoon precipitation and streamflow

Detecting and quantifying the presence of human-induced climate change in regional hydrology is important for studying the impacts of such changes on the water resources systems as well as for reliable future projections and policy making for adaptation. In this article a formal fingerprint-based detection and attribution analysis has been attempted to study the changes in the observed monsoon precipitation and streamflow in the rain-fed Mahanadi River Basin in India, considering the variability across different climate models. This is achieved through the use of observations, several climate model runs, a principal component analysis and regression based statistical downscaling technique, and a Genetic Programming based rainfall-runoff model. It is found that the decreases in observed hydrological variables across the second half of the 20th century lie outside the range that is expected from natural internal variability of climate alone at 95% statistical confidence level, for most of the climate models considered. For several climate models, such changes are consistent with those expected from anthropogenic emissions of greenhouse gases. However, unequivocal attribution to human-induced climate change cannot be claimed across all the climate models and uncertainties in our detection procedure, arising out of various sources including the use of models, cannot be ruled out. Changes in solar irradiance and volcanic activities are considered as other plausible natural external causes of climate change. Time evolution of the anthropogenic climate change ``signal'' in the hydrological observations, above the natural internal climate variability ``noise'' shows that the detection of the signal is achieved earlier in streamflow as compared to precipitation for most of the climate models, suggesting larger impacts of human-induced climate change on streamflow than precipitation at the river basin scale.

Single nucleotide polymorphisms in genes that are common targets of luteotropin and luteolysin in primate corpus luteum: Computational exploration

Luteal insufficiency affects fertility and hence study of mechanisms that regulate corpus luteum (CL) function is of prime importance to overcome infertility problems. Exploration of human genome sequence has helped to study the frequency of single nucleotide polymorphisms (SNPs). Clinical benefits of screening SNPs in infertility are being recognized well in recent times. Examining SNPs in genes associated with maintenance and regression of CL may help to understand unexplained luteal insufficiency and related infertility. Publicly available microarray gene expression databases reveal the global gene expression patterns in primate CL during the different functional state. We intend to explore computationally the deleterious SNPs of human genes reported to be common targets of luteolysin and luteotropin in primate CL Different computational algorithms were used to dissect out the functional significance of SNPs in the luteinizing hormone sensitive genes. The results raise the possibility that screening for SNPs might be integrated to evaluate luteal insufficiency associated with human female infertility for future studies. (C) 2012 Elsevier B.V. All rights reserved,

USHA: unified software and hardware architecture for video decoding

Video decoders used in emerging applications need to be flexible to handle a large variety of video formats and deliver scalable performance to handle wide variations in workloads. In this paper we propose a unified software and hardware architecture for video decoding to achieve scalable performance with flexibility. The light weight processor tiles and the reconfigurable hardware tiles in our architecture enable software and hardware implementations to co-exist, while a programmable interconnect enables dynamic interconnection of the tiles. Our process network oriented compilation flow achieves realization agnostic application partitioning and enables seamless migration across uniprocessor, multi-processor, semi hardware and full hardware implementations of a video decoder. An application quality of service aware scheduler monitors and controls the operation of the entire system. We prove the concept through a prototype of the architecture on an off-the-shelf FPGA. The FPGA prototype shows a scaling in performance from QCIF to 1080p resolutions in four discrete steps. We also demonstrate that the reconfiguration time is short enough to allow migration from one configuration to the other without any frame loss.

The Invertebrate Assemblage Associated with Macrocystis pyrifera and its Utilization as a Food Resource by Kelp Forest Fishes

The present study was designed to examine the following: (1) the taxonomic. spatial, and temporal patterns of availability of all invertebrate species associated with Macrocystis (excluding protozoans and nematodes); (2) the utilization of this invertebrate assemblage as food by kelp forest fishes within the Macrocystis "foliage- searching" feeding guild, as well as proximal mechanisms leading to observed patterns of resource partitioning; and (3) the dynamic relationship between availability and utilization of this food resource. The approach was largely descriptive. with observations collected during a 19-month period from June 1975 to December 1976. Chapter I is an investigation of the resource utilization patterns of four species of kelp forest fishes with respect to food-related resource dimensions. and tests aspects of current theory involving inter- and intraspecific competition. Chapter II is a detailed examination of the invertebrate assemblage associated with Macrocystis and presents life histories of the fishes examined during this study. (PDFs contains 387 pages, chapter 1 is 203 pages, chapter 2 is 184 pages)

Development of appropriateness explicit criteria for cataract extraction by phacoemulsification

Background: Consensus development techniques were used in the late 1980s to create explicit criteria for the appropriateness of cataract extraction. We developed a new appropriateness of indications tool for cataract following the RAND method. We tested the validity of our panel results. Methods: Criteria were developed using a modified Delphi panel judgment process. A panel of 12 ophthalmologists was assembled. Ratings were analyzed regarding the level of agreement among panelists. We studied the influence of all variables on the final panel score using linear and logistic regression models. The explicit criteria developed were summarized by classification and regression tree analysis. Results: Of the 765 indications evaluated by the main panel in the second round, 32.9% were found appropriate, 30.1% uncertain, and 37% inappropriate. Agreement was found in 53% of the indications and disagreement in 0.9%. Seven variables were considered to create the indications and divided into three groups: simple cataract, with diabetic retinopathy, or with other ocular pathologies. The preoperative visual acuity in the cataractous eye and visual function were the variables that best explained the panel scoring. The panel results were synthesized and presented in three decision trees. Misclassification error in the decision trees, as compared with the panel original criteria, was 5.3%. Conclusion: The parameters tested showed acceptable validity for an evaluation tool. These results support the use of this indication algorithm as a screening tool for assessing the appropriateness of cataract extraction in field studies and for the development of practice guidelines.

Geochemistry and resonance ionization of platinum-group elements

Experimental studies were conducted with the goals of 1) determining the origin of Pt- group element (PGE) alloys and associated mineral assemblages in refractory inclusions from meteorites and 2) developing a new ultrasensitive method for the in situ chemical and isotopic analysis of PGE. A general review of the geochemistry and cosmochemistry of the PGE is given, and specific research contributions are presented within the context of this broad framework.

An important step toward understanding the cosmochemistry of the PGE is the determination of the origin of POE-rich metallic phases (most commonly εRu-Fe) that are found in Ca, AJ-rich refractory inclusions (CAI) in C3V meteorites. These metals occur along with γNi-Fe metals, Ni-Fe sulfides and Fe oxides in multiphase opaque assemblages. Laboratory experiments were used to show that the mineral assemblages and textures observed in opaque assemblages could be produced by sulfidation and oxidation of once homogeneous Ni-Fe-PGE metals. Phase equilibria, partitioning and diffusion kinetics were studied in the Ni-Fe-Ru system in order to quantify the conditions of opaque assemblage formation. Phase boundaries and tie lines in the Ni-Fe-Ru system were determined at 1273, 1073 and 873K using an experimental technique that allowed the investigation of a large portion of the Ni-Fe-Ru system with a single experiment at each temperature by establishing a concentration gradient within which local equilibrium between coexisting phases was maintained. A wide miscibility gap was found to be present at each temperature, separating a hexagonal close-packed εRu-Fe phase from a face-centered cubic γNi-Fe phase. Phase equilibria determined here for the Ni-Fe-Ru system, and phase equilibria from the literature for the Ni-Fe-S and Ni-Fe-O systems, were compared with analyses of minerals from opaque assemblages to estimate the temperature and chemical conditions of opaque assemblage formation. It was determined that opaque assemblages equilibrated at a temperature of ~770K, a sulfur fugacity 10 times higher than an equilibrium solar gas, and an oxygen fugacity 10⁶ times higher than an equilibrium solar gas.

Diffusion rates between -γNi-Fe and εRu-Fe metal play a critical role in determining the time (with respect to CAI petrogenesis) and duration of the opaque assemblage equilibration process. The diffusion coefficient for Ru in Ni (D^Ru_Ni) was determined as an analog for the Ni-Fe-Ru system by the thin-film diffusion method in the temperature range of 1073 to 1673K and is given by the expression:

D^Ru_Ni (cm² sec^-1) = 5.0(±0.7) x 10^-3 exp(-2.3(±0.1) x 10¹² erg mole^-1/RT) where R is the gas constant and T is the temperature in K. Based on the rates of dissolution and exsolution of metallic phases in the Ni-Fe-Ru system it is suggested that opaque assemblages equilibrated after the melting and crystallization of host CAI during a metamorphic event of ≥ 10³ years duration. It is inferred that opaque assemblages originated as immiscible metallic liquid droplets in the CAI silicate liquid. The bulk compositions of PGE in these precursor alloys reflects an early stage of condensation from the solar nebula and the partitioning of V between the precursor alloys and CAI silicate liquid reflects the reducing nebular conditions under which CAI were melted. The individual mineral phases now observed in opaque assemblages do not preserve an independent history prior to CAI melting and crystallization, but instead provide important information on the post-accretionary history of C3V meteorites and allow the quantification of the temperature, sulfur fugacity and oxygen fugacity of cooling planetary environments. This contrasts with previous models that called upon the formation of opaque assemblages by aggregation of phases that formed independently under highly variable conditions in the solar nebula prior to the crystallization of CAI.

Analytical studies were carried out on PGE-rich phases from meteorites and the products of synthetic experiments using traditional electron microprobe x-ray analytical techniques. The concentrations of PGE in common minerals from meteorites and terrestrial rocks are far below the ~100 ppm detection limit of the electron microprobe. This has limited the scope of analytical studies to the very few cases where PGE are unusually enriched. To study the distribution of PGE in common minerals will require an in situ analytical technique with much lower detection limits than any methods currently in use. To overcome this limitation, resonance ionization of sputtered atoms was investigated for use as an ultrasensitive in situ analytical technique for the analysis of PGE. The mass spectrometric analysis of Os and Re was investigated using a pulsed primary Ar⁺ ion beam to provide sputtered atoms for resonance ionization mass spectrometry. An ionization scheme for Os that utilizes three resonant energy levels (including an autoionizing energy level) was investigated and found to have superior sensitivity and selectivity compared to nonresonant and one and two energy level resonant ionization schemes. An elemental selectivity for Os over Re of ≥ 10³ was demonstrated. It was found that detuning the ionizing laser from the autoionizing energy level to an arbitrary region in the ionization continuum resulted in a five-fold decrease in signal intensity and a ten-fold decrease in elemental selectivity. Osmium concentrations in synthetic metals and iron meteorites were measured to demonstrate the analytical capabilities of the technique. A linear correlation between Os⁺ signal intensity and the known Os concentration was observed over a range of nearly 10⁴ in Os concentration with an accuracy of ~ ±10%, a millimum detection limit of 7 parts per billion atomic, and a useful yield of 1%. Resonance ionization of sputtered atoms samples the dominant neutral-fraction of sputtered atoms and utilizes multiphoton resonance ionization to achieve high sensitivity and to eliminate atomic and molecular interferences. Matrix effects should be small compared to secondary ion mass spectrometry because ionization occurs in the gas phase and is largely independent of the physical properties of the matrix material. Resonance ionization of sputtered atoms can be applied to in situ chemical analysis of most high ionization potential elements (including all of the PGE) in a wide range of natural and synthetic materials. The high useful yield and elemental selectivity of this method should eventually allow the in situ measurement of Os isotope ratios in some natural samples and in sample extracts enriched in PGE by fire assay fusion.

Phase equilibria and diffusion experiments have provided the basis for a reinterpretation of the origin of opaque assemblages in CAI and have yielded quantitative information on conditions in the primitive solar nebula and cooling planetary environments. Development of the method of resonance ionization of sputtered atoms for the analysis of Os has shown that this technique has wide applications in geochemistry and will for the first time allow in situ studies of the distribution of PGE at the low concentration levels at which they occur in common minerals.

Assessing trends in the density of Atlantic croaker (Micropogonias undulatus): a comparison of passive acoustic and trawl methods

Using data collected simultaneously from a trawl and a hydrophone, we found that temporal and spatial trends in densities of juvenile Atlantic croaker (Micropogonias undulatus) in the Neuse River estuary in North Carolina can be identified by monitoring their sound production. Multivariate analysis of covariance (MA NCOVA) revealed that catch per unit of effort (CPUE) of Atlantic croaker had a significant relationship with the dependent variables of sound level and peak frequency of Atlantic croaker calls. Tests of between-subject correspondence failed to detect relationships between CPUE and either of the call parameters, but statistical power was low. Williamson’s index of spatial overlap indicated that call detection rate (expressed by a 0–3 calling index) was correlated in time and space with Atlantic croaker CPUE. The correspondence between acoustic parameters and trawl catch rates varied by month and by habitat. In general, the calling index had a higher degree of overlap with this species’ density than did the received sound level of their calls. Classification and regression tree analysis identified calling index as the strongest correlate of CPUE. Passive acoustics has the potential to be an inexpensive means of identifying spatial and temporal trends in abundance for soniferous fish species.

Climate signals in the lower stratosphere and upper troposphere

Empirical orthogonal function (EOF) analysis and regression analysis are used to investigate zonally averaged seasonal temperature anomaly patterns and trends in the lower stratosphere and upper troposphere. The first four EOFs explain 64 percent of the temperature variance and can be related, respectively, to the solar flux (SF) and El Niño/Southern Oscillation (ENSO), to the quasi-biennial oscillation (QBO), to atmospheric carbon dioxide (CO2) and turbidity (TB), and to ENSO. The signal of the fourth EOF is modulated in January to March by the solar flux, with the sense of the modulation determined by the phase of the quasi-biennial oscillation.

Effect of water temperature on the growth performance and digestive enzyme activities of Chinese longsnout catfish (Leiocassis longirostris Gunther)

The present study was carried out to investigate the influence of water temperature on the growth performance and digestive enzyme (pepsin, trypsin and lipase) activities of Chinese longsnout catfish. Triplicate groups of Chinese longsnout catfish (35.6 +/- 0.48 g, mean +/- SE) were reared at different water temperatures (20, 24, 28 and 32 degrees C). The feeding rate (FR), specific growth rate (SGR) and feed efficiency ratio (FER) were significantly affected by water temperatures and regression relationships between water temperature and FI, SGR as well as FER were expressed as FR=-0.016T2+0.91T-10.88 (n=12, R2=0.8752), SGR=-0.026T2+1.39T-17.29 (n=12, R2=0.7599) and FER=-0.013T2+0.70T-8.43 (n=12, R2=0.7272). Based on these, the optimum temperatures for FR, SGR and FER were 27.66, 26.69 and 26.44 degrees C respectively. The specific activities of digestive enzymes at 24 or 28 degrees C were significantly higher than that at 20 or 32 degrees C. In addition, there was a significant linear regression between FR or SGR and specific activities of pepsin and lipase, which indicated that pepsin and lipase played important roles in regulating growth through nutrient digestion in Chinese longsnout catfish.

Coalition and distribution in fuzzy dynamic supply chain systems

In this paper, two models of coalition and income's distribution in FSCS (fuzzy supply chain systems) are proposed based on the fuzzy set theory and fuzzy cooperative game theory. The fuzzy dynamic coalition choice's recursive equations are constructed in terms of sup-t composition of fuzzy relations, where t is a triangular norm. The existence of the fuzzy relations in FSCS is also proved. On the other hand, the approaches to ascertain the fuzzy coalition through the choice's recursive equations and distribute the fuzzy income in FSCS by the fuzzy Shapley values are also given. These models are discussed in two parts: the fuzzy dynamic coalition choice of different units in FSCS; the fuzzy income's distribution model among different participators in the same coalition. Furthermore, numerical examples are given aiming at illustrating these models., and the results show that these models are feasible and validity in FSCS.