Toward breast cancer survivability prediction models through improving training space

Due to the difficulties of outlier and skewed data, the prediction of breast cancer survivability has presented many challenges in the field of data mining and pattern precognition, especially in medical research. To solve these problems, we have proposed a hybrid approach to generating higher quality data sets in the creation of improved breast cancer survival prediction models. This approach comprises two main steps: (1) utilization of an outlier filtering approach based on C-Support Vector Classification (C-SVC) to identify and eliminate outlier instances; and (2) application of an over-sampling approach using over-sampling with replacement to increase the number of instances in the minority class. In order to assess the capability and effectiveness of the proposed approach, several measurement methods including basic performance (e.g., accuracy, sensitivity, and specificity), Area Under the receiver operating characteristic Curve (AUC) and F-measure were utilized. Moreover, a 10-fold cross-validation method was used to reduce the bias and variance of the results of breast cancer survivability prediction models. Results have indicated that the proposed approach leads to improving the performance of breast cancer survivability prediction models by up to 28.34% due to the improved training data space.     

Influences of pretreatment and drying methods on composition,micro/molecular structures and some health‐related functional characteristics of dietary fibre powder from orange pulp residues

Influences of pretreatment methods (washing, blanching, soaking in ethanol), drying methods (hot air‐drying, infrared‐drying, microwave‐drying) and drying temperature on selected characteristics of dietary fibre powder from orange pulp residues were investigated. Pretreatments improved the hydration properties, oil‐holding capacity and glucose dialysis retardation index, but led to losses of fibres, decreased α‐amylase inhibitory activity and aflatoxin B₁ adsorption capacity. Hydration properties, oil‐holding capacity and glucose uptake depended on the structure and porosity of the fibre, while the ability to inhibit α‐amylase and AFB₁ adsorption depended more on the content and molecular fingerprinting of the fibre. Drying condition did not significantly affect functional properties of the fibre, except for the ability to uptake glucose, inhibit α‐amylase and adsorb AFB₁ (P ≤ 0.05). Hot air‐drying at 60 °C resulted in fibre with the highest ability to decrease the rate of glucose and AFB₁ adsorption.     

Synthesis of molecularly imprinted polymers from AnAc for the separation of γ-oryzanol

The selectivity of gamma-oryzanol (γ-oryzanol) was recognized by molecularly imprinted polymer (MIP). Polymeric materials were successfully synthesized via thermal polymerization method using γ-oryzanol as template, anacardic acid (AnAc) as functional monomer, toluene as porogen, benzoyl peroxide (BPO) as initiator and divinylbenzene as crosslinker. Binding performance of MIPs was evaluated by MINITAP 14 for variance of analysis, linear regression analysis and adequating model through full factorial experimental technique in terms of adsorption capacity. Analysis of variance with 95% confidence level suggested significant interaction effect (amounts of template, porogen, crosslinker) on adsorption capacity of MIPs. The strongest interaction is between the amount of porogen and the amount of crosslinker. It was also found that a linear regression model for adsorption capacity represents the experimental data with the correlation coefficients (R²) greater than 0.9. The MIP synthesis with 0.8 mmol of template, 6 ml of porogen and 10 ml of crosslinker provided the highest adsorption capacity of MIP (1.14 mg/g-adsorbent). The proposed method is relatively rapid and easy to perform for the separation of γ-oryzanol in non-aqueous systems.     

Amplification of Periodic Temperature Disturbances in a Packed‐Bed Reactor: CO Oxidation over a CuO/Al2O3 Catalyst

Amplification of periodic variations of input temperature in a product‐inhibited reaction — CO oxidation over CuO‐γ‐Al₂O₃ — was investigated experimentally in an insulated packed‐bed reactor. At steady state the temperature profile was elongated compared with that of a reactant‐inhibited CO oxidation over Pt/Al₂O₃, studied elsewhere. Under periodic operation, amplitudes of the resulting travelling temperature waves, monitored downstream from the reaction front, were amplified to a greater extent than those in the reactant‐inhibited CO oxidation over Pt/Al₂O₃. The magnitude of the amplification depended on the perturbation frequency and showed resonance behaviour. The magnitude decreased monotonically with increasing perturbation amplitude.     

On the dynamical instability of packed-bed reactors in the presence of catalyst deactivation

Due to the thermal instability of the packed-bed reactor running an exothermic reaction, unsteady-state operation (for example a fluctuating inflow temperature) can result in a variety of thermal responses. These include the amplification of input temperature perturbations and high-temperature pre-extinction waves. Catalyst deactivation adds further dynamical features to these scenarios. We explore them numerically, using a first-order exothermic reaction and a pseudo-homogeneous (single phase) model of the PBR together with a first-order deactivation model of the catalyst. At low deactivation rate, moving hot spots are found, as well as a non-uniform activity profile of the catalyst. At high deactivation rate, however, high-temperature waves (so-called pre-extinction waves) are followed by the complete extinction of the reactor. The amplification of input temperature perturbations is generally enhanced by the presence of catalyst deactivation. Finally, a power-law model is derived numerically that predicts the resonance frequency for amplification as a function of operating parameters.     

Enhanced Droplet Size Control in Liquid‐Liquid Emulsions Obtained in a Wire‐Guided X‐Mixer

The droplet size in a liquid‐liquid emulsion can be controlled by placing a metal wire along the centerline of an X‐mixer. Droplets gradually form when flowing along the wire, with droplet separation occurring at the tip of the wire rather than at the channel intersection in the X‐mixer. The droplet size is now defined by the Plateau‐Rayleigh instability developing in the axisymmetric annular flow region rather than by a sophisticated and hardly predictable three‐dimensional flow at the channel intersection. The wire‐guided droplet formation allows for fine control of the droplet size by changing the wire diameter, the position of the wire tip, and the flow rates. Further control of the droplet size can be achieved by adjusting the surface tension by adding a surfactant.     

Analysis of Steady‐State Temperature Profiles from a Laboratory Reactor by Means of a Process Simulator

A process simulator was used for the analysis of steady‐state results from a laboratory‐scale tubular reactor for the oxidation of carbon monoxide over a platinum catalyst. From a set of 14 steady‐state experiments, temperature profiles were simulated with two adjustable parameters recovered by optimizing the fit: k°, the pre‐exponential portion of the rate constant, and h_out, the outer wall heat transfer coefficient for the reactor tube. Simulation showed that despite elaborate insulation the reactor did not behave adiabatically. Simulation also predicted fairly well the magnitude of phenomena such as ignition, extinction, and rate hysteresis (caused by changes in feed temperatures or concentrations) but at temperatures below the experimental values.     

Dynamic Analysis of Thermal Behaviour of a Laboratory Reactor Using a Process Simulator

A process simulator was used to study dynamic results from a laboratory‐scale tubular reactor for the oxidation of carbon monoxide (CO) with oxygen over Pt/alumina catalyst. Passive response experiments were used to obtain lumped values of the heat capacity and the overall heat transfer coefficient. Temperature profile experiments were used to obtain the only adjustable kinetic parameter: k°, the pre‐exponential portion of the rate constant. Integral reactor phenomena were studied: step‐decreases in the feed temperature, step‐changes in feed concentration, and ignition of the reactor by pressure. Simulated results provided good agreement with experiments.