No description available.

No description available.

Abrupt decreases and increases in systolic arterial blood pressure produce baroreflex mediated shortening and lengthening, respectively, of the R-R interval. This phenomenon, otherwise known as the cardivagal baroreflex, is best described by the sigmoid relationship between R-R interval length and systolic blood pressure. The linear portion of this relationship is used to derive the slope or gain of the cardiovagal baroreflex. Importantly, lower levels of cardiovagal baroreflex have been associated with poor orthostatic tolerance and an increased cardiovascular disease-related mortality. The most commonly used and accepted technique to assess cardiovagal barorelex gain is the modified Oxford techinique. Bolus injections of sodium nitroprusside followed by phenylephrine HCL are used to decrease and raise blood pressure ~15 mmHg, respectively. The baroreflex control of the cardiac vagal outflow can then be assessed by the relation of the R-R interval to systolic blood pressure. However, the modified Oxford technique does not always reveal the nonlinear nature of baroreflex relations. The reasons for this has been unclear. Thus, analysis of baroreflex gain when nonlinearities are not revealed is problematic. Five classifications of baroreflex trials have been identified: acceptable, threshold-heavy, saturation-heavy, linear-heavy, and random trials. A new method of gain estimation was developed that combines the strengths of the current methods of gain estimation with the knowledge of the classifications of baroreflex trials. Using this method, cardiovagal baroreflex gain assessment can be maximized if threshold-heavy, saturation-heavy, and random trials are filtered out of the analysis and the manual method is used to estimate gain on the remaining trials. In addition, a link seems to exist between the variability of delta and the variability in baroreflex gain between different subjects. / Master of Science

Precis som de flesta företag så har True Software Scandinavia AB en produkt där ute i marknaden. Produkten är en mobil applikation som går under namnet Truecaller. För att kunna leverera den bästa upplevelsen till slut användarna så krävs en stor global telefonkatalog. Förutom att företaget har en statistik på hur det har gått och går för deras produkt så finns även behovet att veta hur det kommer att gå i framtiden. Alltså man vill veta hur stor potential det finns i marknaden och med avseende på detta försöka estimera när man når sitt uppsatta mål. Syftet med detta projekt har varit att skapa ett verktyg som kan samla in data och utföra nödvändiga behandlingar för att kunna estimera den framtida tillväxten. Verktyget skapas med hjälp av programmeringsspråket Java och kommer vara i form av ett skrivbords applikation. Eftersom ett sådant verktyg är stort och komplicerad, så kan det lösas genom att delas upp i mindre bestånds delar. Varje del utvecklas för sig själv och allt sätts ihop på slutet. För att hela verktyget ska kunna vara funktionellt och ge något resultat så måste varje delkomponent fungera korrekt. I resultatet kommer vi att verifiera att varje delkomponent utför sitt arbete felfri. Förutom resultatet så kommer dessutom vissa diskussioner kring generell barhet, kritik och förbättringar på arbetet att tas upp i denna rapport. / Like most companies, True Software Scandinavia AB has a product out in the market. The product is a mobile application, called Truecaller. In order to deliver the best experience to the end users, a large global phonebook is required. In addition that the company has statistics of how things have gone and are going for their product, there is also a need to know how it will go in the future. Thus they want to know how much potential there is in the market and with regard to this, attempt to estimate when they reach their goal. The purpose of this project is to create a tool that can collect data and perform the necessary process in order to estimate future growth. The tool is created using the Java programming language and will be in the form of a desktop application. As these kinds of tools are large and complex, it will be solved by subdividing it into smaller constituent parts. After each part is separately developed, all parts will be put together at the end. In order for the tool to be functional and give any result, each component must work correctly. In the results, we will verify that each component performs its work faultless. Apart from the result, even some discussions about general sustainability, criticisms and improvements on the work will be included in this report.

This thesis presents the use of a new sigmoid activation function in backpropagation artificial neural networks (ANNs). ANNs using conventional activation functions may generalize poorly when trained on a set which includes quirky, mislabeled, unbalanced, or otherwise complicated data. This new activation function is an attempt to improve generalization and reduce overtraining on mislabeled or irrelevant data by restricting training when inputs to the hidden neurons are sufficiently small. This activation function includes a flattened, low-training region which grows or shrinks during back-propagation to ensure a desired proportion of inputs inside the low-training region. With a desired low-training proportion of 0, this activation function reduces to a standard sigmoidal curve. A network with the new activation function implemented in the hidden layer is trained on benchmark data sets and compared with the standard activation function in an attempt to improve area under the curve for the receiver operating characteristic in biological and other classification tasks.

The aims of the studies was to evaluate the short and long term effects on the development of parastomal hernia and stoma complications of a prophylactic prosthetic mesh placed in a sublay position at the index operation.  Also the purpose was to validate a definition of parastomal hernia at clinical examination and a method and a definition of parastomal hernia at CT-scan. In the first two studies 27 patients were randomized to a conventional stoma or to a stoma with the addition of a partly absorbable low weight large pore mesh in sublay position. Patients were examined after one and five years.  After five years the rate of parastomal hernia was 80% with a conventional stoma and 14% with the addition of a mesh.  A prophylactic mesh did not increase the rate of complications. In the third study a prophylactic mesh was intended at stoma formation in 93 consecutive patients in routine surgery. In 75 patients provided with a mesh the rate of parastomal hernia after one year was 13%. Complication rates were not increased in 19 severely contaminated wounds. In the fourth study 27 patients with ostomies were examined by tree surgeons and parastomal hernia was defined as any protrusion in the vicinity of the stoma. CT-scans with patients examined in the supine and prone positions were assessed by three radiologists. Herniation was then defined as any intra abdominal content protruding beyond peritoneum or the presence of a hernia sac. Kappa was 0.85 for surgeons and 0.85 for radiologists with CT-scan in the prone position. Kappa was 0.80 for surgeons and radiologists collectively, with CT-scan in the prone position. Four parastomal hernias detected at CT-scan in the prone position could not be detected in the supine position. A parastomal hernia diagnosed at clinical examination was always detected at CT-scan in the prone position. Conclusions: A prophylactic mesh placed in a sublay position at the index operation reduces the rate of parastomal hernia without increasing the rate of complications. Parastomal hernia should at clinical examination be defined as any protrusion in the vicinity of the stoma with the patient straining in the supine and erect positions.  At CT-scan, with the patient examined in the prone position, herniation should be defined as any intra abdominal content protruding beyond peritoneum or the presence of a hernia sac. / Embargo, publiceras 2011-05-01

Colonoscopy provides a minimally invasive tool for examining and treating the colon without surgery, but current endoscope designs still cause a degree of pain and injury to the colon wall. The most common colonoscopies are long tubes inserted through the rectum, with locomotion actuators, fiber optic lights, cameras, and biopsy tools on the distal end. The stiffness required to support these tools makes it difficult for the scopes to navigate the twisted path of the colon without damaging the inside wall of the colon or distorting its shape. In addition, little is known about how sharp and forceful endoscopes can be without accidentally cutting into tissue during navigation. In order to solve the requirements of stiffness (to support tools) and flexibility (to navigate turns), we expanded on a design by Zehel et al. [49], who proposed surrounding a flexible endoscope with an external exoskeleton structure, with controllable stiffness. The exoskeleton structure is comprised of rigid, articulating tubular units, which are stiffened or relaxed by four control cables. The stiffened or locked exoskeleton structure aids navigation and provides stability for the endoscope when it protrudes beyond the exoskeleton structure for examination and procedures. This research determined the design requirements of such an exoskeleton structure and simulated its behavior in a sigmoid colon model. To predict just how pointed an endoscope can be without damaging tissue under a given force, we extrapolated a strength model of the descending colon from published stress-strain curves of human colon tissue. Next we analyzed how friction, cable forces, and unit angles interact to hold the exoskeleton structure in a locked position. By creating two- and three-dimensional models of the exoskeleton structure, we optimized the dimensions of the units of an exoskeleton structure (diameter, thickness, and leg angle) and cable holders ( cable attachment location) to achieve the turns of the sigmoid colon, while still remaining lockable. Models also predicted the loss of force over the exoskeleton structure due to curving, further determining the required cable angles and friction between units. Finally we determined how the stiffness of the endoscope stem affected locking ability and wear inside the exoskeleton structure. / Ph. D.

<p>Artificial intelligence (AI) is already a fundamental component of computer games. In this context is emotions a growing part in simulating real life. The proposed emotional state module, provides a way for the game agents to select an action in real-time virtual environments. The modules function has been tested with the open-source strategy game ORTS. This thesis proposes a new approach for the design of an interacting network, similar to a spreading activation system, of emotional states that keeps track of emotion intensities changing and interacting over time. The network of emotions can represent any number of persisting states, such as moods, emotions and drives. Any emotional signal can affect every state positively or negatively. The states' response to emotional signals are influenced by the other states represented in the network. The network is contained within an emotional state module. This interactions between emotions are not the focus of much research, neither is the representation model. The focus tend to be on the mechanisms eliciting emotions and on how to express the emotions.</p>

Artificial intelligence (AI) is already a fundamental component of computer games. In this context is emotions a growing part in simulating real life. The proposed emotional state module, provides a way for the game agents to select an action in real-time virtual environments. The modules function has been tested with the open-source strategy game ORTS. This thesis proposes a new approach for the design of an interacting network, similar to a spreading activation system, of emotional states that keeps track of emotion intensities changing and interacting over time. The network of emotions can represent any number of persisting states, such as moods, emotions and drives. Any emotional signal can affect every state positively or negatively. The states' response to emotional signals are influenced by the other states represented in the network. The network is contained within an emotional state module. This interactions between emotions are not the focus of much research, neither is the representation model. The focus tend to be on the mechanisms eliciting emotions and on how to express the emotions.

This Master’s Thesis consists of description of current routing protocols and routers, basic principles of neural networks and their interpretation in connection with the use for routing in data networks and telecommunications networks. In the thesis I focused on neural networks, which use energetic functions to find solution stabled states and their use for data routing. I produced the application software to test and find suitable variables for each function. This application counts the shortest path and is able to change variables to reach the best solution of stabled state of neural network. These solutions are compared with other functions that are usually used in nowadays systems for data network routing.