Extremal properties of certain risk models

Cette thèse s'intéresse à étudier les propriétés extrémales de certains modèles de risque d'intérêt dans diverses applications de l'assurance, de la finance et des statistiques. Cette thèse se développe selon deux axes principaux, à savoir: Dans la première partie, nous nous concentrons sur deux modèles de risques univariés, c'est-à- dire, un modèle de risque de déflation et un modèle de risque de réassurance. Nous étudions le développement des queues de distribution sous certaines conditions des risques commun¬s. Les principaux résultats sont ainsi illustrés par des exemples typiques et des simulations numériques. Enfin, les résultats sont appliqués aux domaines des assurances, par exemple, les approximations de Value-at-Risk, d'espérance conditionnelle unilatérale etc. La deuxième partie de cette thèse est consacrée à trois modèles à deux variables: Le premier modèle concerne la censure à deux variables des événements extrême. Pour ce modèle, nous proposons tout d'abord une classe d'estimateurs pour les coefficients de dépendance et la probabilité des queues de distributions. Ces estimateurs sont flexibles en raison d'un paramètre de réglage. Leurs distributions asymptotiques sont obtenues sous certaines condi¬tions lentes bivariées de second ordre. Ensuite, nous donnons quelques exemples et présentons une petite étude de simulations de Monte Carlo, suivie par une application sur un ensemble de données réelles d'assurance. L'objectif de notre deuxième modèle de risque à deux variables est l'étude de coefficients de dépendance des queues de distributions obliques et asymétriques à deux variables. Ces distri¬butions obliques et asymétriques sont largement utiles dans les applications statistiques. Elles sont générées principalement par le mélange moyenne-variance de lois normales et le mélange de lois normales asymétriques d'échelles, qui distinguent la structure de dépendance de queue comme indiqué par nos principaux résultats. Le troisième modèle de risque à deux variables concerne le rapprochement des maxima de séries triangulaires elliptiques obliques. Les résultats théoriques sont fondés sur certaines hypothèses concernant le périmètre aléatoire sous-jacent des queues de distributions. -- This thesis aims to investigate the extremal properties of certain risk models of interest in vari¬ous applications from insurance, finance and statistics. This thesis develops along two principal lines, namely: In the first part, we focus on two univariate risk models, i.e., deflated risk and reinsurance risk models. Therein we investigate their tail expansions under certain tail conditions of the common risks. Our main results are illustrated by some typical examples and numerical simu¬lations as well. Finally, the findings are formulated into some applications in insurance fields, for instance, the approximations of Value-at-Risk, conditional tail expectations etc. The second part of this thesis is devoted to the following three bivariate models: The first model is concerned with bivariate censoring of extreme events. For this model, we first propose a class of estimators for both tail dependence coefficient and tail probability. These estimators are flexible due to a tuning parameter and their asymptotic distributions are obtained under some second order bivariate slowly varying conditions of the model. Then, we give some examples and present a small Monte Carlo simulation study followed by an application on a real-data set from insurance. The objective of our second bivariate risk model is the investigation of tail dependence coefficient of bivariate skew slash distributions. Such skew slash distributions are extensively useful in statistical applications and they are generated mainly by normal mean-variance mixture and scaled skew-normal mixture, which distinguish the tail dependence structure as shown by our principle results. The third bivariate risk model is concerned with the approximation of the component-wise maxima of skew elliptical triangular arrays. The theoretical results are based on certain tail assumptions on the underlying random radius.

IVA in addition to BMD can change the osteoporosis management in 25% of clinical routine patients

Introduction: Vertebral fracture is one of the major osteoporotic fractures which are unfortunately very often undetected. In addition, it is well known that prevalent vertebral fracture increases dramatically the risk of future additional fracture. Instant Vertebral Assessment (IVA) has been introduced in DXA device couple years ago to ease the detection of such fracture when routine DXA are performed. To correctly use such tool, ISCD provided clinical recommendation on when and how to use it. The aim of our study was to evaluate the ISCD guidelines in clinical routine patients and see how often it may change of patient management. Methods: During two months (March and April 2010), a medical questionnaire was systematically given to our clinical routine patient to check the validity of ISCD IVA recommendations in our population. In addition, all women had BMD measurement at AP spine, Femur and 1/3 radius using a Discovery A System (Hologic, Waltham, USA). When appropriate, IVA measurement had been performed on the same DXA system and had been centrally evaluated by two trained Doctors for fracture status according to the semi-quantitative method of Genant. The reading had been performed when possible between L5 and T4. Results: Out of 210 women seen in the consultation, 109 (52%) of them (mean age 68.2 ± 11.5 years) fulfilled the necessary criteria to have an IVA measurement. Out of these 109 women, 43 (incidence 39.4%) had osteoporosis at one of the three skeletal sites and 31 (incidence 28.4%) had at least one vertebral fracture. 14.7% of women had both osteoporosis and at least one vertebral fracture classifying them as "severe osteoporosis" while 46.8% did not have osteoporosis nor vertebral fracture. 24.8% of the women had osteoporosis but no vertebral fracture while 13.8% of women did have osteoporosis and vertebral fracture (clinical osteoporosis). Conclusion: In conclusion, in 52% of our patients, IVA was needed according to ISCD criteria. In half of them the IVA test influenced of patient management either by changing the type of treatment of simply by classifying patient as "clinical osteoporosis". IVA appears to be an important tool in clinical routine but unfortunately is not yet very often used in most of the centers.

Assessing protein stability of the dimeric DNA-binding domain of E2 human papillomavirus 18 with molecular dynamics

The objective of this study is to understand the structural flexibility and curvature of the E2 protein of human papillomavirus type 18 using molecular dynamics (6 ns). E2 is required for viral DNA replication and its disruption could be an anti-viral strategy. E2 is a dimer, with each monomer folding into a stable open-faced β-sandwich. We calculated the mobility of the E2 dimer and found that it was asymmetric. These different mobilities of E2 monomers suggest that drugs or vaccines could be targeted to the interface between the two monomers.

IVA in addition to BMD can change the osteoporosis management in 25% of clinical routine patients.

Vertebral fracture is one of the major osteoporotic fractures which are unfortunately very often undetected. In addition, it is well known that prevalent vertebral fracture increases dramatically the risk of future additional fracture. Instant Vertebral Assessment (IVA) has been introduced in DXA device couple years ago to ease the detection of such fracture when routine DXA are performed. To correctly use such tool, ISCD provided clinical recommendation on when and how to use it. The aim of our study was to evaluate the ISCD guidelines in clinical routine patients and see how often it may change of patient management. During two months (March and April 2010), a medical questionnaire was systematically given to our clinical routine patient to check the validity of ISCD IVA recommendations in our population. In addition, all women had BMD measurement at AP spine, Femur and 1/3 radius using a Discovery A System (Hologic, Waltham, USA). When appropriate, IVA measurement had been performed on the same DXA system and had been centrally evaluated by two trained Doctors for fracture status according to the semi-quantitative method of Genant. The reading had been performed when possible between L5 and T4. Out of 210 women seen in the consultation, 109 (52%) of them (mean age 68.2±11.5 years) fulfilled the necessary criteria to have an IVA measurement. Out of these 109 women, 43 (incidence 39.4%) had osteoporosis at one of the three skeletal sites and 31 (incidence 28.4%) had at least one vertebral fracture. 14.7% of women had both osteoporosis and at least one vertebral fracture classifying them as "severe osteoporosis" while 46.8% did not have osteoporosis not vertebral fracture. 24.8% of the women had osteoporosis but no vertebral fracture while 13.8% of women did have osteoporosis but vertebral fracture (Clinical osteoporosis). In conclusion, in 52% of our patients, IVA was needed according to ISCD criteria. In half of them the IVA test influenced of patient management either my changing the type of treatment of simply by classifying patient as "clinical osteoporosis". IVA appears to be an important tool in clinical routine but unfortunately is not yet very often use in most of the centers.

Prokaryotic cells: structural organisation of the cytoskeleton and organelles

For many years, prokaryotic cells were distinguished from eukaryotic cells based on the simplicity of their cytoplasm, in which the presence of organelles and cytoskeletal structures had not been discovered. Based on current knowledge, this review describes the complex components of the prokaryotic cell cytoskeleton, including (i) tubulin homologues composed of FtsZ, BtuA, BtuB and several associated proteins, which play a fundamental role in cell division, (ii) actin-like homologues, such as MreB and Mb1, which are involved in controlling cell width and cell length, and (iii) intermediate filament homologues, including crescentin and CfpA, which localise on the concave side of a bacterium and along its inner curvature and associate with its membrane. Some prokaryotes exhibit specialised membrane-bound organelles in the cytoplasm, such as magnetosomes and acidocalcisomes, as well as protein complexes, such as carboxysomes. This review also examines recent data on the presence of nanotubes, which are structures that are well characterised in mammalian cells that allow direct contact and communication between cells.

Light-induced phosphorylation of a membrane protein plays an early role in signal transduction for phototropism in Arabidopsis thaliana.

Blue light is known to cause rapid phosphorylation of a membrane protein in etiolated seedlings of several plant species, a protein that, at least in etiolated pea seedlings and maize coleoptiles, has been shown to be associated with the plasma membrane. The light-driven phosphorylation has been proposed on the basis of correlative evidence to be an early step in the signal transduction chain for phototropism. In the Arabidopsis thaliana mutant JK224, the sensitivity to blue light for induction of first positive phototropism is known to be 20- to 30-fold lower than in wild type, whereas second positive curvature appears to be normal. While light-induced phosphorylation can be demonstrated in crude membrane preparations from shoots of the mutant, the level of phosphorylation is dramatically lower than in wild type, as is the sensitivity to blue light. Another A. thaliana mutant, JK218, that completely lacks any phototropic responses to up to 2 h of irradiation, shows a normal level of light-induced phosphorylation at saturation. Since its gravitropic sensitivity is normal, it is presumably blocked in some step between photoreception and the confluence of the signal transduction pathways for phototropism and gravitropism. We conclude from mutant JK224 that light-induced phosphorylation plays an early role in the signal transduction chain for phototropism in higher plants.

Tenseness of Riemannian flows

We show that any transversally complete Riemannian foliation &em&F&/em& of dimension one on any possibly non-compact manifold M is tense; namely, (M,&em&F&/em&) admits a Riemannian metric such that the mean curvature form of &em&F&/em& is basic. This is a partial generalization of a result of Domínguez, which says that any Riemannian foliation on any compact manifold is tense. Our proof is based on some results of Molino and Sergiescu, and it is simpler than the original proof by Domínguez. As an application, we generalize some well known results including Masa's characterization of tautness.

IVA in addition to BMD can change the osteoporosis management in 25% of clinical routine patients

Introduction Vertebral fracture is one of the major osteoporoticfractures which are unfortunately very often undetected. In addition,it is well known that prevalent vertebral fracture increases dramaticallythe risk of future additional fracture. Instant Vertebral Assessment(IVA) has been introduced in DXA device a couple of years ago toease the detection of such fracture when routine DXA are performed.To correctly use such tool, ISCD provided clinical recommendationon when and how to use it. The aim of our study was to evaluate theISCD guidelines in clinical routine patients and see how often itmay change of patient management.Methods During two months (March and April 2010), a medicalquestionnaire was systematically given to our clinical routine patientto check the validity of ISCD IVA recommendations in our population.In addition, all women had BMD measurement at AP spine,femur and 1/3 radius using a Discovery A System (Hologic, Waltham,USA). When appropriate, IVA measurement had been performedon the same DXA system and had been centrally evaluated by twotrained doctors for fracture status according to the semi-quantitativemethod of Genant. The reading had been performed when possiblebetween L5 and T4.Results Out of 210 women seen in the consultation, 109 (52 %)of them (mean age 68.2 ± 11.5 years) fulfilled the necessary criteriato have an IVA measurement. Out of these 109 women, 43 (incidence39.4 %) had osteoporosis at one of the three skeletal sitesand 31 (incidence 28.4 %) had at least one vertebral fracture. 14.7 %of women had both osteoporosis and at least one vertebral fractureclassifying them as "severe osteoporosis" while 46.8 % did not haveosteoporosis and no vertebral fracture. 24.8 % of the women hadosteoporosis but no vertebral fracture while 13.8 % of women didhave osteoporosis but vertebral fracture (clinical osteoporosis).Conclusions In 52 % of our patients, IVA was needed accordingto ISCD criteria. In half of them the IVA test influenced of patientmanagement either may changing the type of treatment of simplyby classifying patient as "clinical osteoporosis". IVA appears to bean important tool in clinical routine but unfortunately is not yetvery often use in most of the centers.

Large fluctuations in the disassembly rate of microtubules revealed by atomic force microscopy.

Atomic force microscopy (AFM) in situ has been used to observe the cold disassembly dynamics of microtubules at a previously unrealised spatial resolution. Microtubules either electrostatically or covalently bound to aminosilane surfaces disassembled at room temperature under buffer solutions with no free tubulin present. This process was followed by taking sequential tapping-mode AFM images and measuring the change in the microtubule end position as a function of time, with an spatial accuracy down to +/-20nm and a temporal accuracy of +/-1s. As well as giving average disassembly rates on the order of 1-10 tubulin monomers per second, large fluctuations in the disassembly rate were revealed, indicating that the process is far from smooth and linear under these experimental conditions. The surface bound rates measured here are comparable to the rates for GMPCPP-tubulin microtubules free in solution, suggesting that inhibition of tubulin curvature through steric hindrance controls the average, relatively low disassembly rate. The large fluctuations in this rate are thought to be due to multiple pathways in the kinetics of disassembly with differing rate constants and/or stalling due to defects in the microtubule lattice. Microtubules that were covalently bound to the surface left behind the protofilaments covalently cross-linked to the aminosilane via glutaraldehyde during the disassembly process. Further work is needed to quantitatively assess the effects of surface binding on protofibril disassembly rates, reveal any differences in disassembly rates between the plus and minus ends and to enable assembly as well as disassembly to be imaged in the microscope fluid cell in real-time.

Beneficial effects of strontium ranelate compared to alendronate on bone micro-structure - a 2-year study

Aims: In a head-to-head study, we compared the effects of strontium ranelate (SrRan) and alendronate (ALN), anti-osteoporotic agents with antifracture efficacy, on bone microstructure, a component of bone quality, hence of bone strength. Methods: In a randomised, double-dummy, double-blind controlled trial, 88 postmenopausal osteoporotic women were randomised to SrRan 2g/day or ALN 70mg/week for 2 years. Microstructure of the distal radius and distal tibia were assessed by HR-pQCT after 3,6,12,18 and 24 months of treatment. Primary endpoint was HR-pQCT variables relative changes from baseline. An ITT analysis was applied. Results: Baseline characteristics were similar in both groups (mean ±SD): age: 63.6±7.5 vs. 63.7±7.6 yrs; L1-L4T Score: -2.7±0.8 vs. -2.8±0.8g/cm², Cortical Thickness (CTh), trabecular bone fraction (BV/TV) and cortical density=721±242 vs. 753±263μm, 9.5±2.5 vs. 9.3±2.7%, and 750±87 vs. 745±78mg/cm3 respectively. Over 2 yrs, distal radius values changes were within 1 to 2% without significant differences except cortical density. In contrast distal tibia CTh, BV/TV, trabecular and cortical densities increased significantly more in the SrRan group than in the ALN group (Table). No significant between-group differences were observed for the remaining measured parameter (trabecular number, trabecular spacing, and trabecular thickness). After 2 years, L1- L4 and hip aBMD increases were similar to results from pivotal trials (L1-L4:+6.5% and +5.6%;total hip:+4.1% and +2.9%, in Sr- Ran and ALN groups, respectively). In the SrRan group, bALP increased by a median of 18% (p<0.001) and sCTX decreased by a median of -16% (p=0.005) while in the ALN group, bALP and CTX decreased by median of -31% (p<0.001) and -59% (p<0.001) respectively. Relative changes from baseline to last observation (%) SrRan ALN Estimated between group difference p value CTh (μm) 6.29±9.53 0.93±6.23 5.411±1.836 0.004 BV/TV (%) 2.48±5.13 0.84±3.81 1.783±0.852 0.040 Trabecular density (mgHA/cm3) 2.47±5.07 0.88±4.00 1.729±0.859 0.048 Cortical density (mgHA/cm3) 1.43±2.77 0.36±2.14 1.137±0.530 0.045 The two treatments were well tolerated. Conclusions: Within the constraints related to HRpQCT technology, it appears that strontium ranelate has greater effects than alendronate on distal tibia cortical thickness, trabecular and cortical bone densities in women with postmenopausal osteoporosis after two years of treatment. A concomitant significant increase in bone formation marker is observed in the SrRan group.

Are patterns of bone loss in anorexic and postmenopausal women similar? Preliminary results using high resolution peripheral computed tomography.

This study intended to compare bone density and architecture in three groups of women: young women with anorexia nervosa (AN), an age-matched control group of young women, and healthy late postmenopausal women. Three-dimensional peripheral quantitative high resolution computed-tomography (HR-pQCT) at the ultradistal radius, a technology providing measures of cortical and trabecular bone density and microarchitecture, was performed in the three cohorts. Thirty-six women with AN aged 18-30years (mean duration of AN: 5.8years), 83 healthy late postmenopausal women aged 70-81 as well as 30 age-matched healthy young women were assessed. The overall cortical and trabecular bone density (D100), the absolute thickness of the cortical bone (CTh), and the absolute number of trabecules per area (TbN) were significantly lower in AN patients compared with healthy young women. The absolute number of trabecules per area (TbN) in AN and postmenopausal women was similar, but significantly lower than in healthy young women. The comparison between AN patients and post-menopausal women is of interest because the latter reach bone peak mass around the middle of the fertile age span whereas the former usually lose bone before reaching optimal bone density and structure. This study shows that bone mineral density and bone compacta thickness in AN are lower than those in controls but still higher than those in postmenopause. Bone compacta density in AN is similar as in controls. However, bone inner structure in AN is degraded to a similar extent as in postmenopause. This last finding is particularly troubling.

The AO comprehensive classification of pediatric long-bone fractures: a web-based multicenter agreement study.

The first AO comprehensive pediatric long-bone fracture classification system has been proposed following a structured path of development and validation with experienced pediatric surgeons. A Web-based multicenter agreement study involving 70 surgeons in 15 clinics and 5 countries was conducted to assess the reliability and accuracy of this classification when used by a wide range of surgeons with various levels of experience. Training was provided at each clinic before the session. Using the Internet, participants could log in at any time and classify 275 supracondylar, radius, and tibia fractures at their own pace. The fracture diagnosis was made following the hierarchy of the classification system using both clinical terminology and codes. kappa coefficients for the single-surgeon diagnosis of epiphyseal, metaphyseal, or diaphyseal fracture type were 0.66, 0.80, and 0.91, respectively. Median accuracy estimates for each bone and type were all greater than 80%. Depending on their experience and specialization, surgeons greatly varied in their ability to classify fractures. Pediatric training and at least 2 years of experience were associated with significant improvement in reliability and accuracy. Kappa coefficients for diagnosis of specific child patterns were 0.51, 0.63, and 0.48 for epiphyseal, metaphyseal, and diaphyseal fractures, respectively. Identified reasons for coding discrepancies were related to different understandings of terminology and definitions, as well as poor quality radiographic images. Results supported some minor adjustments in the coding of fracture type and child patterns. This classification system received wide acceptance and support among the surgeons involved. As long as appropriate training could be performed, the system classification was reliable, especially among surgeons with a minimum of 2 years of clinical experience. We encourage broad-based consultation between surgeons' international societies and the use of this classification system in the context of clinical practice as well as prospectively for clinical studies.

Greenstick fractures of the middle third of the forearm. A prospective multi-centre study.

BACKGROUND: Greenstick fractures suffered during growth have a high risk for refracture and posttraumatic deformity, particularly at the forearm diaphysis. The use of a preemptive completion of the fracture by manipulation of the concave cortex is controversial and data supporting this approach are few. AIM: Aim of this study was to determine the factors which predispose to refracture and deformities, and to define therapeutic strategies. METHODS: We prospectively gathered clinical and radiographic data over a period of one year on greenstick fractures of the middle third of the forearm in children as part of a multi-centre study. Endpoint was a follow-up visit at one year. Radiographic deformity, state of consolidation at resumption of physical activities and refracture rate were analysed statistically (ANOVA, Student's t-test and Pearson's chi-square test) with regard to patient age, gender, fracture type, therapy and time in plaster. RESULTS: We collected the data of 103 patients (63 boys, 40 girls), average age 6.6 years (1.3-14.5 years), the vast majority of whom had a combined greenstick fracture of the radius and ulna. 6.7% of the patients sustained a refracture within 49 days (29-76) after plaster removal. They were significantly older (p=0.017) with a significantly higher incidence of manual completion of the fracture with radiographic signs of partial consolidation (p=0.025). Residual deformities were significantly smaller after completion of the fracture compared to reduction without completion (p=0.019) or plaster fixation alone (p<0.005). CONCLUSIONS: Completion of a greenstick fracture does not prevent refracture. Nevertheless, it diminishes the extent of secondary deformities in cases where the primary angulation exceeds the remodelling capacity. Prevention of refracture should include a routine radiographic follow-up 4-6 weeks after injury with continuation of plaster fixation in cases of partial consolidation.

Alterations of bone microarchitecture in young patients with inflammatory bowel diseases are associated with fracture risk during growth

Aims: Inflammatory bowel diseases (IBD) appearing during childhood and adolescence compromise peak bone mass acquisition and increase fracture risk. The structural determinants of bone fragility in IBD however remain unknown. Methods: We investigated volumetric bone mineral density (vBMD), trabecular and cortical bone microstructure at distal radius and tibia by high-resolution pQCT (XtremeCT, Scanco, Switzerland), aBMD at distal radius, hip and spine and vertebral fracture assessment (VFA) by DXA in 107 young patients (mean age 22.8 yrs, range 12.2-33.7 yrs; 62 females and 45 males) with Crohn's disease (n=75), ulcerative colitis (n=25), undetermined colitis (n=2), and no definitive diagnosis (n=5), and in 389 healthy young individuals. Results: Mean disease duration was 6.1 yrs, 89/107 IBD patients received corticosteroids, 83 other immunomodulators, and 59 vitamin D. Clinical fractures were reported by 38 patients (mean age at 1st fracture, 12.6 yrs), the vast majority of the forearm, arm or hand; 5 had vertebral crush fractures (Grade 1 or 2) and 11 had multiple fractures. As compared to healthy controls (matched 2:1 for age, sex, height and fracture history), the 102 patients with established IBD had similar weight but significantly lower aBMD at all sites, lower trabecular (Tb) BV/TV and number, and greater Tb separation and inhomogeneous Tb distribution (1/SD TbN) at both distal radius and tibia, lower tibia cortical thickness (CTh), but no differences in cortical vBMD nor bone perimeter. Among IBD's, aBMD was not associated with fractures (by logistic regression adjusted for age, age square, sex, height, weight and protein intake). However, radius and tibia Tb BV/TV, thickness and SD 1/TbN, as well as radius Tb separation and perimeter, were significantly associated with fracture risk (fully adjusted as above), whereas cortical vBMD and CTh were not. After adjustment for aBMD at radius, respectively at femur neck, radius SD 1/TbN and tibia BV/TV, TbTh and perimeter remained independently associated with fracture risk. Conclusions: Young subjects with IBD have low bone mass and poor bone microarchitecture compared to healthy controls. Alterations of bone microarchitecture, particularly in the trabecular bone compartment, are specifically associated with increased fracture risk during growth.

Rapid prototyping of compliant human aortic roots for assessment of valved stents.

Adequate in-vitro training in valved stents deployment as well as testing of the latter devices requires compliant real-size models of the human aortic root. The casting methods utilized up to now are multi-step, time consuming and complicated. We pursued a goal of building a flexible 3D model in a single-step procedure. We created a precise 3D CAD model of a human aortic root using previously published anatomical and geometrical data and printed it using a novel rapid prototyping system developed by the Fab@Home project. As a material for 3D fabrication we used common house-hold silicone and afterwards dip-coated several models with dispersion silicone one or two times. To assess the production precision we compared the size of the final product with the CAD model. Compliance of the models was measured and compared with native porcine aortic root. Total fabrication time was 3 h and 20 min. Dip-coating one or two times with dispersion silicone if applied took one or two extra days, respectively. The error in dimensions of non-coated aortic root model compared to the CAD design was <3.0% along X, Y-axes and 4.1% along Z-axis. Compliance of a non-coated model as judged by the changes of radius values in the radial direction by 16.39% is significantly different (P<0.001) from native aortic tissue--23.54% at the pressure of 80-100 mmHg. Rapid prototyping of compliant, life-size anatomical models with the Fab@Home 3D printer is feasible--it is very quick compared to previous casting methods.