Class of 2010 Abstract / OBJECTIVES: The purpose of this study was to demonstrate that an increase in INR is associated with a hemorrhagic event in patients taking the oral anticoagulant warfarin. METHODS: A retrospective review of data from 18 patients previously stable on warfarin therapy with an elevation in INR at the time of a hemorrhagic event. Patients were receiving warfarin treatment in the anticoagulation clinic at the Southern Arizona VA Healthcare system from April 2008 to December 2009. Primary outcome measures included a comparison of INR, warfarin dose, and hematocrit at baseline, within 7 days of the event, and during follow-­‐up. RESULTS: A significant increase in INR was observed from baseline to the event (2.5 +/-­‐ 0.36 vs 6.2 +/-­‐ 3.2; p = 0.0002) but differences in INR during all periods of follow-­‐up did not differ from baseline (p = 0.35 – 0.99). When compared with baseline, differences in warfarin dose reached statistical significance when all 12 weeks of follow-­‐up were included (34.4 +/-­‐ 13.8 mg vs 32.4+/-­‐ 15.5 mg; p = 0.01) but were not significant when only the last 8 weeks (p = 0.06) or 4 weeks (p = 0.16) were included. Hematocrit values decreased significantly following hemorrhage (39.8 +/-­‐ 3.63 vs 33.5 +/-­‐ 5.72; p = 0.0002) before trending toward baseline (39.85 +/-­‐ 3.63 vs 37.13 +/-­‐ 4.72; p = 0.007). CONCLUSIONS: Hemorrhagic events were associated with an increased INR in previously stable warfarin patients. The mean weekly warfarin dose required to maintain a therapeutic INR returned to baseline within 8 weeks of the hemorrhagic event.

Flexural mechanism design is one of the latest field for mechanism researches in these recent years. In this paper, it will address one method of normalized dimensional parameters. According to the traditionally change of the appearance of flexural hinge, the different parameters will be designed through to the differences of several combination. On the other hand, we use the Finite Element Method to analyze the kinematics behavior of different hinges. Finally, we come up with a series of complete data that provide designers a reference for designing.

Dans cette thèse, nous nous intéressons à la classification des images de textures avec aucune connaissance a priori sur les conditions de numérisation. Cette classification selon des définitions pré-établies de matériaux repose sur des algorithmes qui extraient des descripteurs visuels.A cette fin, nous introduisons tout d'abord une variante de descripteurs par motifs binaires locaux (Local Binary Patterns).Dans cette proposition, une approche statistique est suivie pour représenter les textures statiques.Elle incorpore la quantité d'information complémentaire des niveaux de gris des images dans des opérateurs basés LBP.Nous avons nommé cette nouvelle méthode "Completed Local Entropy Binary Patterns (CLEBP)".CLEBP capture la distribution des relations entre les mesures statistiques des données aléatoires d'une image, l'ensemble étant calculé pour tous les pixels au sein d'une structure locale.Sans la moindre étape préalable d'apprentissage, ni de calibration automatique, les descriptions CLEBP contiennent à la fois des informations locales et globales des textures, tout en étant robustes aux variations externes.En outre, nous utilisons le filtrage inspiré par la biologie, ou biologically-inspired filtering (BF), qui simule la rétine humaine via une phase de prétraitement.Nous montrons que notre approche est complémentaire avec les LBP conventionnels, et les deux combinés offrent de meilleurs résultats que l'une des deux méthodes seule.Les résultats expérimentaux sur quatre bases de texture, Outex, KTH-TIPS-2b, CURet, et UIUC montrent que notre approche est plus performante que les méthodes actuelles.Nous introduisons également un cadre formel basé sur une combinaison de descripteurs pour la classification de textures.Au sein de ce cadre, nous combinons des descripteurs LBP invariants en rotation et en échelle, et de faible dimension, avec les réseaux de dispersion, ou scattering networks (ScatNet).Les résultats expérimentaux montrent que l'approche proposée est capable d'extraire des descripteurs riches à de nombreuses orientations et échelles.Les textures sont modélisées par une concaténation des codes LBP et valeurs moyennes des coefficients ScatNet.Nous proposons également d'utiliser le filtrage inspiré par la biologie, ou biologically-inspired filtering (BF), pour améliorer la resistance des descripteurs LBP.Nous démontrons par l'expérience que ces nouveaux descripteurs présentent de meilleurs résultats que les approches usuelles de l'état de l'art.Ces résultats sont obtenus sur des bases réelles qui contiennent de nombreuses avec des variations significatives.Nous proposons aussi un nouveau réseau conçu par l'expertise appelé réseaux de convolution normalisée, ou normalized convolution network.Celui-ci est inspiré du modèle des ScatNet, auquel deux modifications ont été apportées.La première repose sur l'utilisation de la convolution normalisé en lieu et place de la convolution standard.La deuxième propose de remplacer le calcul de la valeur moyenne des coefficients du réseaux par une agrégation avec la méthode des vecteurs de Fisher.Les expériences montrent des résultats compétitifs sur de nombreuses bases de textures.Enfin, tout au long de cette thèse, nous avons montré par l'expérience qu'il est possible d'obtenir de très bons résultats de classification en utilisant des techniques peu coûteuses en ressources. / This thesis investigates the complementary approaches for classifying texture images.The thesis begins by proposing a Local Binary Pattern (LBP) variant for efficient texture classification.In this proposed method, a statistical approach to static texture representation is developed. It incorporates the complementary quantity information of image intensity into the LBP-based operators. We name our LBP variant `the completed local entropy binary patterns (CLEBP)'. CLEBP captures the distribution of the relationships between statistical measures of image data randomness, calculated over all pixels within a local structure. Without any pre-learning process and any additional parameters to be learned, the CLEBP descriptors convey both global and local information about texture while being robust to external variations. Furthermore, we use biologically-inspired filtering (BF) which simulates the performance of human retina as preprocessing technique. It is shown that our approach and the conventional LBP have the complementary strength and that by combining these algorithms, one obtains better results than either of them considered separately. Experimental results on four large texture databases show that our approach is more efficient than contemporary ones.We then introduce a framework which is a feature combination approach to the problem of texture classification. In this framework, we combine Local Binary Pattern (LBP) features with low dimensional, rotation and scale invariant counterparts, the handcrafted scattering network (ScatNet). The experimental results show that the proposed approach is capable of extracting rich features at multiple orientations and scales. Textures are modeled by concatenating histogram of LBP codes and the mean values of ScatNet coefficients. Then, we propose using Biological Inspired Filtering (BF) preprocessing technique to enhance the robustness of LBP features. We have demonstrated by experiment that the novel features extracted from the proposed framework achieve superior performance as compared to their traditional counterparts when benchmarked on real-world databases containing many classes with significant imaging variations.In addition, we propose a novel handcrafted network called normalized convolution network. It is inspired by the model of ScatNet with two important modification. Firstly, normalized convolution substitute for standard convolution in ScatNet model to extract richer texture features. Secondly, Instead of using mean values of the network coefficients, Fisher vector is exploited as an aggregation method. Experiments show that our proposed network gains competitive classification results on many difficult texture benchmarks.Finally, throughout the thesis, we have proved by experiments that the proposed approaches gain good classification results with low resource required.

Alluvial groundwater systems are an important source of water for communities and biodiverse riparian corridors throughout the arid and semi-arid Basin and Range Geological Province of western North America. These aquifers and their attendant desert streams have been depleted to support a growing population, while projected climate change could lead to more extreme episodes of drought and precipitation in the future. The only source of replenishment to these aquifers is recharge. This dissertation builds upon previous work to characterize and quantify recharge in arid and semi-arid basins by characterizing the intra-annual seasonality of recharge across the Basin and Range Province, and considering how climate change might impact recharge seasonality and volume, as well as fragile riparian corridors that depend on these hydrologic processes. First, the seasonality of recharge in a basin in the sparsely-studied southern extent of the Basin and Range Province is determined using stable water isotopes of seasonal precipitation and groundwater, and geochemical signatures of groundwater and surface water. In northwestern Mexico in the southern reaches of the Basin and Range, recharge is dominated by winter precipitation (69% ± 42%) and occurs primarily in the uplands. Second, isotopically-based estimates of seasonal recharge fractions in basins across the region are compared to identify patterns in recharge seasonality, and used to evaluate a simple water budget-based model for estimating recharge seasonality, the normalized seasonal wetness index (NSWI). Winter precipitation makes up the majority of annual recharge throughout the region, and North American Monsoon (NAM) precipitation has a disproportionately weak impact on recharge. The NSWI does well in estimating recharge seasonality for basins in the northern Basin and Range, but less so in basins that experience NAM precipitation. Third, the seasonal variation in riparian and non-riparian vegetation greenness, represented by the normalized difference vegetation index (NDVI), is characterized in several of the study basins and climatic and hydrologic controls are identified. Temperature was the most significant driver of vegetation greenness, but precipitation and recharge seasonality played a significant role in some basins at some elevations. Major contributions of this work include a better understanding of recharge in a monsoon-dominated basin, the characterization of recharge seasonality at a regional scale, evaluation of an estimation method for recharge seasonality, and an interpretation of the interaction of seasonal hydrologic processes, vegetation dynamics, and climate change.

Class of 2010 Abstract / OBJECTIVES: To compare the INR-based results of pharmacist anticoagulation management services to non-pharmacist managed anticoagulation therapy. METHODS: A meta-analysis was performed using studies that reported on pharmacists’ services and interventions in anticoagulation. Two reviewers independently assessed each record retrieved by the literature search, and studies were eliminated if both reviewers agreed that the study did not meet the inclusion criteria. The analysis found eight studies that were eligible for inclusion. The primary independent variable was the presence or absence of pharmacist services. The primary dependent variables were the proportion of patients within INR range, the duration of time within that range, and the time required to achieve that range. Extracted data were pooled and entered into the meta-analysis, and a forest plot was constructed. The a-priori alpha level was 0.05. RESULTS: The eight studies were divided into two groups: The INR group consisted of four studies which used INR test data points as the unit of analysis. The Patient group included four studies which measured the patient’s INR only at the end of the study. The INR group demonstrated a significant difference favoring pharmacist services, with a p-value of 0.02 for the group. There was no significant difference between pharmacist services and the control group with respect to the Patient group. When all eight studies were pooled together, the overall forest plot did demonstrate significant favorability for pharmacist services, with an odds ratio of 1.58, indicating that pharmacist provided services improved care by 58% (Z = 2.8;p p = 0.01). CONCLUSIONS: The advantage of a pharmacist service in reaching goal INR was significant compared to the control group of non- pharmacist care.

Student Number : 0216831W - MSc research report - School of Environmental Sciences - Faculty of Science / This report deals with the relationship between the NDVI obtained from the NOAA AVHRR sensor and that obtained from the MODIS sensor. The relationship is quantitatively assessed for distinct polygons over various land-cover types in the northeastern Kwa-Zulu Natal Province of South Africa. Spatial and temporal variations in the relationships are addressed and discussed with reference to spectral response, sunsensor- target geometries and atmospheric factors. Specifically, various methods are investigated to estimate a MODIS-equivalent NDVI from the AVHRR NDVI and in so doing create the potential to develop a self-consistent NDVI between the historically available AVHRR NDVI dataset and the relatively new MODIS NDVI dataset. NOAA-16 AVHRR NDVI data and AQUA MODIS NDVI data for the two-year period from January 2002 to December 2003 are used to develop the method. A linear relationship exists between the AVHRR and MODIS NDVI. However, spatial variations in the relationship in terms of land-cover and mean NDVI are pointed out. The potential of atmospheric corrections applied to AVHRR data through a radiative transfer atmospheric correction model to improve the relationship between the two NDVI datasets is also investigated. The importance of geo-location accuracy of the AVHRR NDVI dataset is assessed in the light of the accuracy obtainable with the proposed method to estimate a MODIS-equivalent NDVI from the AVHRR NDVI. A method to estimate the MODIS NDVI from the AVHRR NDVI that takes the mean AVHRR NDVI value into account, as opposed to a constant linear relationship over all the points, is proposed. Atmospheric correction is shown not to improve the accuracy of the method in a statistically significant way. The root-mean-square error of the proposed method is in the order of 0.05 NDVI units and varies between 0.5 and 2 standard deviations of the MODIS NDVI over an entire season.

Image segmentation is an important task in computer vision and understanding. Graph Cuts have been shown to be useful in image segmentation problems. Using a criterion for segmentation optimality, they can obtain segmentation without relying heavily on a priori information regarding the specific type of object. Discussed here are a few approximations to the Normalized Cut criterion, the solving of which has been shown to be an NP-hard problem. Two Normalized Cut algorithms have been previously proposed, and a third is proposed here which accomplishes approximation by a similar method as one of the previous algorithms. It is also more efficient than either of the previously proposed Normalized Cut approximations.

Introduction: It is recommended to place all the vacuum tubes directly on a sample cradle after vein puncture to prevent analytic error. This recommendation is not always easy to follow because the samples are taken by different professionals under different situations.  The three most common analyses, platelets count, haemoglobin and prothrombin time were tested.  Therefore, it was interesting to compare results from the three most common analyses with or without sample cradle, to evaluate the influence of this step on the result. Methods: Three analyses were preformed, using blood from 50 different persons. Each person gave two vacuum tubes, each contained 4.5mL of venous blood for the study. Tubes containing EDTA were used for platelet counts and measurement of haemoglobin and tubes containing citrate were used for prothrombin time-analysis. One of the tubes was placed, as recommended, directly on the sample cradle while the other tube was placed flat on a bench for 10 minutes before it was placed on the sample cradle. Results: There was a clear difference in platelet counts with and without immediate cradling but only minor difference between the results for haemoglobin and International Normalized Ratio. Conclusion: Some analyses seem to be more sensitive for variation in cradling than others. For platelet count it was important to immediately rock the tubes but for determination of prothrombine time and hemoglobin it had a small impact. The small impact on the results is probably due to the efficiency of the anticoagulant in the vacuum tubes.

The Caspian Region includes the Caspian Sea and five littoral states: Azerbaijan, Iran, Turkmenistan, Kazakhstan and Russian. 40% of the Caspian coastal zone is arid, 69% of this territory undergone desertification according to international reports. Among the reasons are soil erosion caused by water, wind and irrigation, the salinization of soil, intense bioresources usage, and soil pollution due to oil extraction and production. Desertification is a serious problem, at global, national and local scales. It is important to know what should be sustained or developed in order to protect land from desertification. The generalization of data over desertification processes in Caspian countries, studying the dynamics of this process in space and time could help facilitate measures to counter regional desertification. To understand Caspian Region coastal desertification phenomenon, vegetation cover satellite images for the years 1982 – 2006 were investigated to give map vegetation changes over time. The Normalized Difference Vegetation Index (NDVI) data for this study was derived from the Global Inventory Modeling and Mapping Studies (GIMMS) dataset, with the spatial resolution of 8 km. A coastal strip 160 km from the coast, divided by countries, was investigated. Theanalyses were focused on extent and severity of vegetation cover degradation, and possible causes such as landscape, land use history and culture, climatic changes and policies. The aim was to address questions related to desertification phenomenon, by focusing on Caspian Region time-series of vegetation cover data and investigation patterns of desertification in the region. In this study evidence of land degradation in the Caspian Region countries was found to occur on local scales or sub-national scales rather than across the regional as a whole. Changes in vegetation cover revealed by AVHRR NDVI appeared to be reversible in character and were dependent on the climate conditions, and anthropogenic impact in approximately equal proportions.

In stochastic representation an image is a sample function of an array of random variables which is called a random field. For characterizing an ensemble of images, we choose an autoregressive model as our image model. An image model often applies to image processing such as image data compression and image restoration. Therefore the validity of the image model affect it¡¦s performance of image processing. The output of the AR model depends on its parameters ¡V system transition matrix and generating noise. Hence the validity of this model is related to these two parameters. How to seek the standard of the validity of the image model is a problem. We exploit performance of image model¡¦s application ¡V image restoration - to find a method of determining the validity of the image model. In our paper we find a relation between image restoration performance and image model¡¦s parameters by the Kalman filtering equations. An image model with lower generating noise power and system transition matrix is better for image restoration and is considered a good image model. In the analysis of the parameters of the image model, we can meet the requirements of the parameters by image segmentation method, residual image method and normalized image method. In addition it also helps us understand the Kalman filter much more and know how to find the solution of similar problems.