Biomedical subjects

Peter York

Publications and source records attributed to Peter York.

21 records · Page 2Linked to original sources

PubMed2002

The effect of experimental design on the modeling of a tablet coating formulation using artificial neural networks.

The aim of this study was to investigate the effect of experimental design strategy on the modeling of a film coating formulation by artificial neural networks (ANNs). Box-Behnken, central composite and pseudo-random designs of 102, 90 and 100 simulated records, respectively were used to train a multilayer perceptron (MLP) ANN comprising six input and two output nodes separated by a single hidden layer of five nodes. Network over-training was limited by using a test set of 40 pseudo-randomly distributed records. The models were validated using a set of 60 pseudo-randomly distributed records. Crack velocity was highly curved with respect to pigment particle size and size distribution. Similarly, film opacity was highly curved in response to pigment concentration and film thickness. The Box-Behnken and central composite designs generated models that were unable to predict crack velocity and showed extensive bias in prediction of film opacity. The pseudo-random design was unable to predict crack velocity of the test data set but yielded acceptable predictions for the validation set. Film opacity was well predicted by the pseudo-random design model. The poor predictive ability of the Box-Behnken and central composite models was attributed to poor interpolation of the high curvature of the response surfaces. In contrast, the pseudo-random design mapped the interior of the design space allowing improved interpolation and predictive ability. It is concluded that Box-Behnken and central composite experimental designs are inappropriate for ANN modeling of highly curved responses and that extensive internal mapping of the design space is essential to generate predictive ANN models.

Drug Design↗

PubMed2002

Influence of polymorphism on the surface energetics of salmeterol xinafoate crystallized from supercritical fluids.

PURPOSE: To characterize the surface thermodynamic properties of two polymorphic forms (I and II) of salmeterol xinafoate (SX) prepared from supercritical fluids and a commercial micronized SX (form 1) sample (MSX). METHODS: Inverse gas chromatographic analysis was conducted on the SX samples at 30, 40, 50, and 60 degrees C using the following probes at infinite dilution: nonpolar probes (NPs; alkane C5-C9 series); and polar probes (PPs; i.e., dichloromethane, chloroform, acetone, ethyl acetate, diethyl ether, and tetrahydrofuran). Surface thermodynamic parameters of adsorption and Hansen solubility parameters were calculated from the retention times of the probes. RESULTS: The free energies of adsorption (- deltaG(A)) of the three samples obtained at various temperatures follow this order: SX-II > MSX approximately/= SX-I for the NPs; and SX-II > MSX > SX-I for the PPs. For both NPs and PPs, SX-II exhibits a less negative enthalpy of adsorption (deltaH(A)) and a much less negative entropy of adsorption (ASA) than MSX and SX-I, suggesting that the high -AGA of SX-II is contributed by a considerably reduced entropy loss. The dispersive component of surface free energy (gammas(D)) is the highest for MSX but the lowest for SX-II at all temperatures studied, whereas the specific component of surface free energy of adsorption (-deltaG(A)SP) is higher for SX-II than for SX-I. That SX-II displays the highest -deltaG(A) for the NP but the lowest gammasD of all the SX samples may be explained by the additional -AGA change associated with an increased mobility of the probe molecules on the less stable and more disordered SX-II surface. The acid and base parameters, K(A) and K(D) that were derived from deltaH(A)SP reveal significant differences in the relative acid and base properties among the samples. The calculated Hansen solubility parameters (deltaD, deltap, and deltaH) indicate that the surface of SX-II is the most polar and most energetic of all the three samples in terms of specific interactions (mostly hydrogen bonding). CONCLUSIONS: The metastable SX-II polymorph possesses a higher surface free energy, higher surface entropy, and a more polar surface than the stable SX-I polymorph.

Albuterol↗

PubMedDate not supplied

Optimisation of the predictive ability of artificial neural network (ANN) models: a comparison of three ANN programs and four classes of training algorithm.

The purpose of this study was to determine whether artificial neural network (ANN) programs implementing different backpropagation algorithms and default settings are capable of generating equivalent highly predictive models. Three ANN packages were used: INForm, CAD/Chem and MATLAB. Twenty variants of gradient descent, conjugate gradient, quasi-Newton and Bayesian regularization algorithms were used to train networks containing a single hidden layer of 3-12 nodes. All INForm and CAD/Chem models trained satisfactorily for tensile strength, disintegration time and percentage dissolution at 15, 30, 45 and 60 min. Similarly, acceptable training was obtained for MATLAB models using Bayesian regularization. Training of MATLAB models with other algorithms was erratic. This effect was attributed to a tendency for the MATLAB implementation of the algorithms to attenuate training in local minima of the error surface. Predictive models for tablet capping and friability could not be generated. The most predictive models from each ANN package varied with respect to the optimum network architecture and training algorithm. No significant differences were found in the predictive ability of these models. It is concluded that comparable models are obtainable from different ANN programs provided that both the network architecture and training algorithm are optimised. A broad strategy for optimisation of the predictive ability of an ANN model is proposed.

Algorithms↗