This thesis includes the development, construction and testing of internal organ phantoms, with focus on the liver, for biomechanical testing. Phantoms have various biomedical applications such as surgical simulations, minimally invasive surgery, soft tissue characterization, diagnostic tools and instrumentation calibration. However, there is little work present in literature regarding phantoms and the work that is currently available does not account for the non-linear viscoelastic properties as well as the Glisson’s capsule. In this work, three different phantoms are presented: a fluid-filled phantom, a perfused phantom and a hydrogel-based liver phantom. A testing apparatus is designed, built and used to measure the force-displacement data during the indentation of the phantom. The first phantom that is designed and constructed follows the basis of a fluid-filled vessel. It is composed of a linear low-density polyethylene (LLDPE) bag filled with different fluids namely: water, a 1:1 water/glycerine mixture and glycerine. The phantoms are subjected to quasi-static loading as well as relaxation testing. The effect of density and viscosity, its size, and confined and unconfined boundary conditions are characterized. The second phantom is designed to investigate the effects of hepatic macrocirculation on the biomechanical properties of the liver. The phantom is made of two-part silicone (Smooth-On, ECOFLEX 00-30), and contains a network of conduits to model the large ! iv! blood vessels in the liver. A perfusion system that captures the general features of the human hepatic circulation is used to help investigate the effects of the different flow parameters such as pressure and flow rate on the biomechanical characteristics of the liver. The perfusion system is designed to reproduce comparable pressures to the human portal vein and hepatic artery. The third phantom is made of two parts, a hydrogel inner layer with a LLDPE outer layer. The idea behind this phantom is to represent the organ as accurately as possible by accounting for the capsule that surrounds the organ as well as the biphasic (solid and fluid) nature of the organ. A biphasic poroviscoelastic model is used to model the hydrogel while the LLDPE uses a non-linear hyperelastic and viscoelastic model. Modeling is done in ABAQUS to fit the experimental data obtained from quasi-static indentation and relaxation testing using a parametric study. In conclusion, phantoms replicating the non-linear viscoelastic properties observed in organs are presented and characterized. Main Thesis Contributions • Development and characterization of a simple fluid-filled phantom to represent the mechanical properties of the liver • Development and characterization of hydrogel-based liver phantom with representation of the biphasic nature of the organ and the Glisson’s capsule. ! v! • Development and characterization of perfused liver phantom with ability to be recreated with various vessel configurations. • Development of testing set-up to characterize various phantoms.

The effects of supplementation of riboflavin along with Bacillus subtilis was investigated on growth performance and intestinal health of broilers under coccidial challenge. Treatments are 3 × 2 × 2 factorial arrangement, which includes 3 levels of riboflavin (0.75, 6.6 (recommended), and 20 ppm), with or without Bacillus subtilis, and with or without coccidial challenge. Coccidial challenge impaired the intestinal morphology and reduced body weight (BW), body weight gain (BWG), and feed intake and increased FCR between d 14-28. However, the cocci impact on birds was reduced along with age, as the effects on internal organs was reduced, which allowed birds to recover. However, coccidiosis reduced overall BW and BWG and increased abdominal fat pad weight and slight woody breast incidence. In conclusion, riboflavin at tested level was unable to enhance growth performance and intestinal health, but its inclusion altered birds response to coccidial challege other than feeding Bacillus subtlis alone.

CoordenaÃÃo de AperfeÃoamento de Pessoal de NÃvel Superior / Objetivou-se com o presente estudo avaliar o efeito de classe sexual (castrado e nÃo castrado) e nÃveis de restriÃÃo alimentar (0%, 30% e 60%) sobre as caracterÃsticas quantitativas dos componentes carcaÃa e nÃo carcaÃa de cordeiros Santa InÃs. Foram utilizados 30 cordeiros, 15 castrados e 15 nÃo castrados, com aproximadamente dois meses de idade e peso corporal inicial mÃdio de 13 Â 1,49 kg. Os cordeiros foram distribuÃdos em delineamento inteiramente casualizado em arranjo fatorial 3x2 (nÃvel de restriÃÃo x classe sexual), de acordo com a quantidade de alimento fornecida (0%, 30% e 60% de restriÃÃo). A duraÃÃo do experimento foi determinada pelo tempo necessÃrio para que a mÃdia de peso corporal dos animais de um dos tratamentos atingisse 28 kg, ocasiÃo em que todos os 30 animais foram abatidos. Houve influÃncia da classe sexual sobre as caracterÃsticas quantitativas peso corporal ao abate, peso do corpo vazio, peso de carcaÃa quente e peso de carcaÃa fria, sendo os animais nÃo castrados os que apresentaram maiores valores. O peso da perna mostrou-se maior para animais nÃo castrados (1,727) que para os castrados (1,637). Houve interaÃÃo entre classe sexual e nÃveis de restriÃÃo para profundidade do tÃrax e comprimento da perna. Houve efeito linear decrescente de acordo com o aumento dos nÃveis de restriÃÃo alimentar para as caracterÃsticas e cortes de carcaÃa, exceto para rendimento de perna e lombo posterior. NÃo foi observada interaÃÃo entre nÃveis de restriÃÃo e classe sexual para peso dos componentes nÃo carcaÃa. Houve diferenÃa entre classes sexuais para os pesos dos rins, patas, sangue e intestino grosso de animais nÃo castrados, os quais obtiveram mÃdias superiores aos animais castrados. Para sangue, cabeÃa, pele, pata, rÃmen-retÃculo, abomaso, intestino delgado, intestino grosso e gorduras perirrenal, omental, mesentÃrica e do coraÃÃo houve efeito linear decrescente para os nÃveis de restriÃÃo. As caracterÃsticas quantitativas dos componentes carcaÃa e nÃo carcaÃa de cordeiros Santa InÃs sÃo influenciadas pelas classes sexuais e nÃveis restriÃÃo alimentar. / The present study was conducted to investigate the effect of sex class(castrated and non- castrated) and levels of feed restriction (0%, 30% e 60%) on the quantitative characteristics of carcass and non-carcass components of Santa Ines lambs. We used thirty Santa Ines lambs, castrated (n=15) and nonâcastrated (n=15) with approximately two months of age and average initial body weight of 13 Â 1.49 kg. The lambs were assigned to a completely randomized design in a 3x2 factorial arrangement (restriction level x sex class), according to the amount of food provided (0%, 30% and 60% restriction). Experiment period was determined based in the time required for the animals of one of the groups achieved 28 kg of average body weight, when all 30 animals were slaughtered. We found the influence of sex class on quantitative traits body weight at slaughter, empty body weight, hot and cold carcass weight non castrated animals showed higher averages than the castrated animals. The weight of the leg of the non- castred animal (1,727) was biggest than the castrated animals (1,637). Was observed interaction between sex class and dietary restriction for the all of characteristics of carcass, except to leg yield and back rib. Interaction between restriction levels and genders was not observed for non-carcass components. There were differences between genders for the weights of the kidneys, paws, blood and large intestine of non-castrated animals, which had higher averages than castrated animals. A linear decreasing effect to the restriction levels were observed for blood, head, skin, paw, rumen-reticulum, abomasum, small intestine, large intestine and perirenal fat, omental fat, mesenteric and heart fats. Quantitative characteristics of carcass and non-carcass components of growing Santa Ines lambs are influenced by sex class and dietary restriction levels.