Detection of Optimal Models in Parameter Space with Support Vector Machines

The paper proposes an approach aimed at detecting optimal model parameter combinations to achieve the most representative description of uncertainty in the model performance. A classification problem is posed to find the regions of good fitting models according to the values of a cost function. Support Vector Machine (SVM) classification in the parameter space is applied to decide if a forward model simulation is to be computed for a particular generated model. SVM is particularly designed to tackle classification problems in high-dimensional space in a non-parametric and non-linear way. SVM decision boundaries determine the regions that are subject to the largest uncertainty in the cost function classification, and, therefore, provide guidelines for further iterative exploration of the model space. The proposed approach is illustrated by a synthetic example of fluid flow through porous media, which features highly variable response due to the parameter values' combination.

PEG-b-PPS diblock copolymer aggregates for hydrophobic drug solubilization and release: cyclosporin A as an example

Micelles formed from amphiphilic block copolymers have been explored in recent years as carriers for hydrophobic drugs. In an aqueous environment, the hydrophobic blocks form the core of the micelle, which can host lipophilic drugs, while the hydrophilic blocks form the corona or outer shell and stabilize the interface between the hydrophobic core and the external medium. In the present work, mesophase behavior and drug encapsulation were explored in the AB block copolymeric amphiphile composed of poly(ethylene glycol) (PEG) as a hydrophile and poly(propylene sulfide) PPS as a hydrophobe, using the immunosuppressive drug cyclosporin A (CsA) as an example of a highly hydrophobic drug. Block copolymers with a degree of polymerization of 44 on the PEG and of 10, 20 and 40 on the PPS respectively (abbreviated as PEG44-b-PPS10, PEG44-b-PPS20, PEG44-b-PPS40) were synthesized and characterized. Drug-loaded polymeric micelles were obtained by the cosolvent displacement method as well as the remarkably simple method of dispersing the warm polymer melt, with drug dissolved therein, in warm water. Effective drug solubility up to 2 mg/mL in aqueous media was facilitated by the PEG- b-PPS micelles, with loading levels up to 19% w/w being achieved. Release was burst-free and sustained over periods of 9-12 days. These micelles demonstrate interesting solubilization characteristics, due to the low glass transition temperature, highly hydrophobic nature, and good solvent properties of the PPS block

Seismic attenuation in heterogeneous partially saturated rocks

We study wave-induced fluid flow effects in porous rocks partially saturated with gas and water, where the saturation patterns are governed by mesoscopic heterogeneities associated with the dry frame properties. The link between the dry frame properties and the gas saturation is defined by the assumption of capillary pressure equilibrium, which in the presence of heterogeneity implies that neighboring regions can exhibit different levels of saturation. In order to determine the equivalent attenuation and phase velocity of the synthetic rock samples considered in this study, we apply a numerical upscaling procedure, which permits to take into account mesoscopic heterogeneities associated with the dry frame properties as well as spatially continuous variations of the pore fluid properties. We consider numerical experiments to analyze such effects in heterogeneous partially saturated porous media, where the saturation field is determined by realistic variations in porosity. Our results indicate that the spatially continuous nature of gas saturation inherent to this study is a critical parameter controlling the seismic response of these environments, which in turn suggests that the physical mechanisms governing partial saturation should be accounted for when analyzing seismic data in a poro-elastic context.

Exclusion of lumbar arteries by aortic endovascular grafts: can angiography demonstrate sealing characteristics?

BACKGROUND: This study evaluates sealing characteristics of two designs of endovascular grafts by angiographic demonstration of exclusion of porcine lumbar arteries. METHODS: 6 endovascular grafts (3 self-expandable with integrated polyurethane wall versus 3 nitinol structures covered with polyester fabric) were implanted in 6 porcine aortae. Perfusion of lumbar arteries was assessed by angiography after implantation and by angiography and dissection at graft explantation after 4 +/- 2 months. Tissue healing was evaluated by light and scanning electron microscopy. RESULTS: Immediate exclusion of the lumbar arteries was achieved in 14/31 vessels (12 by polyurethane grafts and 2 by polyester grafts, p < 0.001). Follow-up angiography and dissection at explantation revealed perfusion of 30/31 lumbar arteries with a collateral network in most cases. Another reason for reperfusion of initially excluded branches was distention of the polyurethane grafts with resulting shortening allowing reperfusion of 8 of the 31 originally covered branches. Histological examination revealed a complete neointimal lining and a tight contact between endovascular grafts and aorta. CONCLUSIONS: The immediate angiographic demonstration of exclusion of lumbar arteries predicts sealing characteristics of endovascular grafts. Later angiographic reappearance is due to development of a collateral network and possible shortening of self-expandable devices.