314 resultados para charge-exchange resonances


Relevância:

20.00% 20.00%

Publicador:

Resumo:

The number of patients treated by haemodialysis (HD) is continuously increasing. The complications associated with vascular accesses represent the first cause of hospitalisation in these patients. Since 2001 nephrologists, surgeons, angiologists and radiologists at the CHUV are working to develop a multidisciplinary model that includes planning and monitoring of HD accesses. In this setting the echo-Doppler represents an important tool of investigation. Every patient is discussed and decisions are taken during a weekly multidisciplinary meeting. A network has been created with nephrologists of peripheral centres and other specialists. This model allows to centralize investigational information and coordinate patient care while keeping and even developing some investigational activities and treatment in peripheral centres.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

Polarized tip growth is a fundamental cellular process in many eukaryotic organisms, mediating growth of neuronal axons and dendrites or fungal hyphae. In plants, pollen and root hairs are cellular model systems for analysing tip growth. Cell growth depends on membrane traffic. The regulation of this membrane traffic is largely unknown for tip-growing cells, in contrast to cells exhibiting intercalary growth. Here we show that in Arabidopsis, GBF1-related exchange factors for the ARF GTPases (ARF GEFs) GNOM and GNL2 play essential roles in polar tip growth of root hairs and pollen, respectively. When expressed from the same promoter, GNL2 (in contrast to the early-secretory ARF GEF GNL1) is able to replace GNOM in polar recycling of the auxin efflux regulator PIN1 from endosomes to the basal plasma membrane in non-tip growing cells. Thus, polar recycling facilitates polar tip growth, and GNL2 seems to have evolved to meet the specific requirement of fast-growing pollen in higher plants.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

Abstract The solvability of the problem of fair exchange in a synchronous system subject to Byzantine failures is investigated in this work. The fair exchange problem arises when a group of processes are required to exchange digital items in a fair manner, which means that either each process obtains the item it was expecting or no process obtains any information on, the inputs of others. After introducing a novel specification of fair exchange that clearly separates safety and liveness, we give an overview of the difficulty of solving such a problem in the context of a fully-connected topology. On one hand, we show that no solution to fair exchange exists in the absence of an identified process that every process can trust a priori; on the other, a well-known solution to fair exchange relying on a trusted third party is recalled. These two results lead us to complete our system model with a flexible representation of the notion of trust. We then show that fair exchange is solvable if and only if a connectivity condition, named the reachable majority condition, is satisfied. The necessity of the condition is proven by an impossibility result and its sufficiency by presenting a general solution to fair exchange relying on a set of trusted processes. The focus is then turned towards a specific network topology in order to provide a fully decentralized, yet realistic, solution to fair exchange. The general solution mentioned above is optimized by reducing the computational load assumed by trusted processes as far as possible. Accordingly, our fair exchange protocol relies on trusted tamperproof modules that have limited communication abilities and are only required in key steps of the algorithm. This modular solution is then implemented in the context of a pedagogical application developed for illustrating and apprehending the complexity of fair exchange. This application, which also includes the implementation of a wide range of Byzantine behaviors, allows executions of the algorithm to be set up and monitored through a graphical display. Surprisingly, some of our results on fair exchange seem contradictory with those found in the literature of secure multiparty computation, a problem from the field of modern cryptography, although the two problems have much in common. Both problems are closely related to the notion of trusted third party, but their approaches and descriptions differ greatly. By introducing a common specification framework, a comparison is proposed in order to clarify their differences and the possible origins of the confusion between them. This leads us to introduce the problem of generalized fair computation, a generalization of fair exchange. Finally, a solution to this new problem is given by generalizing our modular solution to fair exchange