Relating formulation variables to in vitro dissolution using an artificial neural network

Fast, Spectroscopy-Based Prediction of In Vitro Dissolution Profile of Extended Release Tablets Using Artificial Neural Networks

The pharmaceutical industry has never seen such a vast development in process analytical methods as in the last decade. The application of near-infrared (NIR) and Raman spectroscopy in monitoring production lines has also become widespread. This work aims to utilize the large amount of information collected by these methods by building an artificial neural network (ANN) model that can predict the dissolution profile of the scanned tablets. An extended release formulation containing drotaverine (DR) as a model drug was developed and tablets were produced with 37 different settings, with the variables being the DR content, the hydroxypropyl methylcellulose (HPMC) content and compression force. NIR and Raman spectra of the tablets were recorded in both the transmission and reflection method. The spectra were used to build a partial least squares prediction model for the DR and HPMC content. The ANN model used these predicted values, along with the measured compression force, as input data. It was found that models based on both NIR and Raman spectra were capable of predicting the dissolution profile of the test tablets within the acceptance limit of the f2 difference factor. The performance of these ANN models was compared to PLS models using the same data as input, and the prediction of the ANN models was found to be more accurate. The proposed method accomplishes the prediction of the dissolution profile of extended release tablets using either NIR or Raman spectra.

Optimization of matrix tablets controlled drug release using Elman dynamic neural networks and decision trees

The main objective of the study was to develop artificial intelligence methods for optimization of drug release from matrix tablets regardless of the matrix type. Static and dynamic artificial neural networks of the same topology were developed to model dissolution profiles of different matrix tablets types (hydrophilic/lipid) using formulation composition, compression force used for tableting and tablets porosity and tensile strength as input data. Potential application of decision trees in discovering knowledge from experimental data was also investigated. Polyethylene oxide polymer and glyceryl palmitostearate were used as matrix forming materials for hydrophilic and lipid matrix tablets, respectively whereas selected model drugs were diclofenac sodium and caffeine. Matrix tablets were prepared by direct compression method and tested for in vitro dissolution profiles. Optimization of static and dynamic neural networks used for modeling of drug release was performed using Monte Carlo simulations or genetic algorithms optimizer. Decision trees were constructed following discretization of data. Calculated difference (f(1)) and similarity (f(2)) factors for predicted and experimentally obtained dissolution profiles of test matrix tablets formulations indicate that Elman dynamic neural networks as well as decision trees are capable of accurate predictions of both hydrophilic and lipid matrix tablets dissolution profiles. Elman neural networks were compared to most frequently used static network, Multi-layered perceptron, and superiority of Elman networks have been demonstrated. Developed methods allow simple, yet very precise way of drug release predictions for both hydrophilic and lipid matrix tablets having controlled drug release.

Novel approaches to neural and evolutionary computing in pharmaceutical formulation: challenges and new possibilities

The development of commercial pharmaceutical formulations can involve extensive experimentation that generates a large amount of data. Understanding such data and discovering the key relationships within them can be a complex process, adding considerably to the time and expense taken to get a product to market. However, new computational techniques such as neural and evolutionary computing have the potential to accelerate the mining and modeling of data, and these methodologies are now being packaged in a way that makes them readily accessible to the product formulator. This article outlines the basis of these technologies and reviews their use in pharmaceutical formulation, showing that they have gained acceptance as practical research tools, especially when integrated with complementary optimization and visualization tools. Neural and evolutionary computing are gaining widespread acceptance in the field of pharmaceutical formulation, with results being comparable with, or better than, those from traditional statistical methods.

Generalization of a prototype intelligent hybrid system for hard gelatin capsule formulation development

The aim of this project was to expand a previously developed prototype expert network for use in the analysis of multiple biopharmaceutics classification system (BCS) class II drugs. The model drugs used were carbamazepine, chlorpropamide, diazepam, ibuprofen, ketoprofen, naproxen, and piroxicam. Recommended formulations were manufactured and tested for dissolution performance. A comprehensive training data set for the model drugs was developed and used to retrain the artificial neural network. The training and the system were validated based on the comparison of predicted and observed performance of the recommended formulations. The initial test of the system resulted in high error values, indicating poor prediction capabilities for drugs other than piroxicam. A new data set, containing 182 batches, was used for retraining. Ten percent of the test batches were used for cross-validation, resulting in models with R2 > or = 70%. The comparison of observed performance to the predicted performance found that the system predicted successfully. The hybrid network was generally able to predict the amount of drug dissolved within 5% for the model drugs. Through validation, the system was proven to be capable of designing formulations that met specific drug performance criteria. By including parameters to address wettability and the intrinsic dissolution characteristics of the drugs, the hybrid system was shown to be suitable for analysis of multiple BCS class II drugs.

Prediction of kinetics of doxorubicin release from sulfopropyl dextran ion-exchange microspheres using artificial neural networks

The purpose of this work was to develop artificial neural networks (ANN) models to predict in vitro release kinetics of doxorubicin (Dox) delivered by sulfopropyl dextran ion-exchange microspheres. Four ANN models for responses at different time points were developed to describe the release profiles of Dox. Model selection was performed using the Akaike information criterion (AIC). Sixteen data sets were used to train the ANN models and two data sets for the validation. Good correlations were obtained between the observed and predicted release profiles for the two randomly selected validation data sets. The difference factor (f1) and similarity factor (f2) between the ANN predicted and the observed release profiles indicated good performance of the ANN models. The established models were then applied to predict release kinetics of Dox from the microspheres of various initial loadings in media of different ionic strengths and NaCl/CaCl2 ratios. The results suggested that ANN offered a flexible and effective approach to predicting the kinetics of Dox release from the ion-exchange microspheres.

Artificial neural network as an alternative to multiple regression analysis in optimizing formulation parameters of cytarabine liposomes

The objective of the study was to optimize the formulation parameters of cytarabine liposomes by using artificial neural networks (ANN) and multiple regression analysis using 3(3) factorial design (FD). As model formulations, 27 formulations were prepared. The formulation variables, drug (cytarabine)/lipid (phosphatidyl choline [PC] and cholesterol [Chol]) molar ratio (X1), PC/Chol in percentage ratio of total lipids (X2), and the volume of hydration medium (X3) were selected as the independent variables; and the percentage drug entrapment (PDE) was selected as the dependent variable. A set of causal factors was used as tutorial data for ANN and fed into a computer. The optimization was performed by minimizing the generalized distance between the predicted values of each response and the optimized one that was obtained individually. In case of 3(3) factorial design, a second-order full-model polynomial equation and a reduced model were established by subjecting the transformed values of independent variables to multiple regression analysis, and contour plots were drawn using the equation. The optimization methods developed by both ANN and FD were validated by preparing another 5 liposomal formulations. The predetermined PDE and the experimental data were compared with predicted data by paired t test, no statistically significant difference was observed. ANN showed less error compared with multiple regression analysis. These findings demonstrate that ANN provides more accurate prediction and is quite useful in the optimization of pharmaceutical formulations when compared with the multiple regression analysis method.

The use of artificial neural networks for the selection of the most appropriate formulation and processing variables in order to predict the in vitro dissolution of sustained release minitablets

The objective of this work was to apply artificial neural networks (ANNs) to examine the relative importance of various factors, both formulation and process, governing the in-vitro dissolution from enteric-coated sustained release (SR) minitablets. Input feature selection (IFS) algorithms were used in order to give an estimate of the relative importance of the various formulation and processing variables in determining minitablet dissolution rate. Both forward and backward stepwise algorithms were used as well as genetic algorithms. Networks were subsequently trained using the back propagation algorithm in order to check whether or not the IFS process had correctly located any unimportant inputs. IFS gave consistent rankings for the importance of the various formulation and processing variables in determining the release of drug from minitablets. Consistent ranking was achieved for both indices of the release process; ie, the time taken for release to commence through the enteric coat (T(lag)) and that for the drug to diffuse through the SR matrix of the minitablet into the dissolution medium (T9(0-10)). In the case of the T(lag) phase, the main coating parameters, along with the original batch blend size and the blend time with lubricant, were found to have most influence. By contrast, with the T(90-10 phase), the amounts of matrix forming polymer and direct compression filler were most important. In the subsequent training of the ANNs, removal of inputs regarded as less important led to improved network performance. ANNs were capable of ranking the relative importance of the various formulations and processing variables that influenced the release rate of the drug from minitablets. This could be done for all main stages of the release process. Subsequent training of the ANN verified that removal of less relevant inputs from the training process led to an improved performance from the ANN.

Comparison of four artificial neural network software programs used to predict the in vitro dissolution of controlled-release tablets

The purpose of this study was to evaluate four commercially available artificial neural network (ANN) software programs: NeuroShell2 v3.0, BrainMaker v3.7, CAD/Chem v5.0, and NeuralWorks Professional II/Plus for prediction of in vitro dissolution-time profiles of controlled-release tablets containing a model sympathomimetic drug. Seven independent formulation variables and three other tablet variables (moisture content of granules, granule particle size, and tablet hardness), for 22 tablet formulations, were used as the ANN model input. In vitro dissolution time-profiles at 10 different sampling times were used as the output. The models' optimum architectures were determined for each ANN software by varying the number of hidden layers and number of nodes in hidden layer(s). The ANN developed from the four software programs were validated by predicting the in vitro dissolution time-profiles of each of the 19 formulations, which were excluded from the training process. Although the same data set was used, the optimum ANN architectures generated from the four software programs were different. Using the four optimum ANN models, the plots of predicted vs. observed percentage of drug dissolved gave slopes ranging from 0.95 to 1.01 and r2 values ranging from 0.95 to 0.99 for all 190 dissolution data points for the 19 training formulations. The difference factors (f1) and similarity factors (f2) between the ANN predicted and the observed in vitro dissolution profiles were also used to compare the predictions for the four software programs. It was concluded that the four programs provided reasonable predictions of in vitro dissolution profiles for the data set employed in this study, with NeuralShell2 showing the best overall prediction.

Effect of varying optimization parameters on optimization by guided evolutionary simulated annealing (GESA) using a tablet film coat as an example formulation

The purpose of this study was to investigate the effect of varying optimization parameters on the proposed optimum of a tablet coating formulation requiring minimization of crack velocity and maximization of film opacity. An artificial neural network (ANN) comprising six input and two output nodes separated by a single hidden layer of five nodes was trained using 100 pseudo-randomly distributed records and optimized by guided evolutionary simulated annealing (GESA). GESA was unable to identify a formulation that satisfied both a crack velocity of 0 ms(-1) and a film opacity of 100% due to conflict centred on the response of the properties to variation in pigment particle size. Constraining film thickness exacerbated the property conflict. By adjusting property weights (i.e. the relative importance of each property), GESA was able to propose formulations that were either crack resistant or that were fully opaque. Reducing the stringency of the performance criteria (crack velocity >0 ms(-1), film opacity <100%) enabled GESA to propose optima that met or exceeded the looser targets. Under these conditions, starting GESA from different locations within model space resulted in the proposal of different optima. Therefore, application of loose targets resulted in the identification of an optimal zone within which all formulations satisfied these less stringent performance criteria. It is concluded that application of the most stringent performance criteria and selection of appropriate property weights is necessary for unequivocal identification of the true optimum. A strategy for optimization experiments is proposed.

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.

Prediction of drug content and hardness of intact tablets using artificial neural network and near-infrared spectroscopy

The purpose of this study was to predict drug content and hardness of intact tablets using artificial neural networks (ANN) and near-infrared spectroscopy (NIRS). Tablets for the drug content study were compressed from mixtures of Avicel PH-101, 0.5% magnesium stearate, and varying concentrations (0%, 1%, 2%, 5%, 10%, 20%, and 40% w/w) of theophylline. Tablets for the hardness study were compressed from mixtures of Avicel PH-101 and 0.5% magnesium stearate at varying compression forces ranging from 0.4 to 1 ton. An Intact Analyzer was used to obtain near infrared spectra from the tablets with varying drug contents, whereas a Rapid Content Analyzer (RCA) was used to obtain spectral data from the tablets with varying hardness. Two sets of tablets from each batch (i.e., tablets with varying drug content and hardness) were randomly selected. One set of tablets was used to generate appropriate calibration models, while the other set was used as the unknown (test) set. A total of 10 ANN calibration models (5 each with 10 and 160 inputs at appropriate wavelengths) and five separate 4-factor partial least squares (PLS) calibration models were generated to predict drug contents of the test tablets from the spectral data. For the prediction of tablet hardness, two ANN calibration models (one each with 10 and 160 inputs) and two 4-factor PLS calibration models were generated and used to predict the hardness of test tablets. The PLS calibration models were generated using Vision software. Prediction of drug contents of test tablets using the ANN calibration models generated with 10 inputs was significantly better than the prediction obtained with the ANN calibration models with 160 inputs. For tablets with low drug concentrations (less than or equal to 2% w/w) prediction of drug content was better with either of the two ANN calibration models than with the PLS calibration models. However, prediction of drug contents of tablets with greater than or equal to 5% w/w drug was better with the PLS calibration models than with the ANN calibration models. Prediction of tablet hardness was better with the ANN calibration models generated with either 10 or 160 inputs than with the PLS calibration models. This work demonstrated that a well-trained ANN model is a powerful alternative technique for analysis of NIRS data. Moreover, the technique could be used in instances when the conventional modeling of data does not work adequately.

Preformulation studies and characterization of the physicochemical properties of amorphous polymers using artificial neural networks

The utility of artificial neural networks (ANNs) as a preformulation tool to determine the physicochemical properties of amorphous polymers such as the hydration characteristics, glass transition temperatures and rheological properties was investigated. The neural network simulator, CAD/Chem, based on the delta back-propagation paradigm was used for this study. The ANNs software was trained with sets of experimental data consisting of different polymer blends with known water-uptake profiles, glass transition temperatures and viscosity values. A set of similar data, not initially exposed to the ANNs was used to validate the ability of the ANNs to recognize patterns. The results of this investigation indicate that the ANNs accurately predicted the water-uptake, glass transition temperatures and viscosities of different amorphous polymers and their physical blends with a low % error (0-8%) of prediction. The ANNs also showed good correlation between the water-uptake and changes in the glass transition temperatures of the polymers. This study demonstrated the potential of the ANNs as a preformulation tool to evaluate the characteristics of amorphous polymers. This is particularly relevant when designing sustained release formulations that require the use of a fast hydrating polymer matrix.

The application of an artificial neural network and pharmacokinetic simulations in the design of controlled-release dosage forms

The objective of this work is to use an artificial neural network (ANN) and pharmacokinetic simulations in the design of controlled-release formulations with predictable in vitro and in vivo behavior. Seven formulation variables and three other tablet variables (moisture, particle size and hardness) for 22 tablet formulations of a model sympathomimetic drug were used as the ANN model input, and in vitro dissolution-time profiles at ten different sampling times were used as output. An ANN model was constructed by selecting the optimal number of iterations and model structure (the number of hidden layers and number of hidden layer nodes). The optimized ANN model was used for prediction of formulations based on two desired target in vitro dissolution-time profiles and two desired bioavailability profiles. For three of the four predicted formulations there was very good agreement between the ANN predicted and the observed in vitro and simulated in vivo properties. This work illustrates the potential for an artificial neural network, along with pharmacokinetic simulations, to assist in the development of complex dosage forms.