Quality by Design Approach: Application of Artificial Intelligence Techniques of Tablets Manufactured by Direct Compression

The publication of the International Conference of Harmonization (ICH) Q8, Q9, and Q10 guidelines paved the way for the standardization of quality after the Food and Drug Administration issued current Good Manufacturing Practices guidelines in 2003. “Quality by Design”, mentioned in the ICH Q8 guideline, offers a better scientific understanding of critical process and product qualities using knowledge obtained during the life cycle of a product. In this scope, the “knowledge space” is a summary of all process knowledge obtained during product development, and the “design space” is the area in which a product can be manufactured within acceptable limits. To create the spaces, artificial neural networks (ANNs) can be used to emphasize the multidimensional interactions of input variables and to closely bind these variables to a design space. This helps guide the experimental design process to include interactions among the input variables, along with modeling and optimization of pharmaceutical formulations. The objective of this study was to develop an integrated multivariate approach to obtain a quality product based on an understanding of the cause–effect relationships between formulation ingredients and product properties with ANNs and genetic programming on the ramipril tablets prepared by the direct compression method. In this study, the data are generated through the systematic application of the design of experiments (DoE) principles and optimization studies using artificial neural networks and neurofuzzy logic programs.KEY WORDS: artificial neural networks (ANNs), gene expression programming (GEP), optimization, quality by design (QbD)     

Size Control in the Nanoprecipitation Process of Stable Iodine (127I) Using Microchannel Reactor—Optimization by Artificial Neural Networks

In this study, nanosuspension of stable iodine (¹²⁷I) was prepared by nanoprecipitation process in microfluidic devices. Then, size of particles was optimized using artificial neural networks (ANNs) modeling. The size of prepared particles was evaluated by dynamic light scattering. The response surfaces obtained from ANNs model illustrated the determining effect of input variables (solvent and antisolvent flow rate, surfactant concentration, and solvent temperature) on the output variable (nanoparticle size). Comparing the 3D graphs revealed that solvent and antisolvent flow rate had reverse relation with size of nanoparticles. Also, those graphs indicated that the solvent temperature at low values had an indirect relation with size of stable iodine (¹²⁷I) nanoparticles, while at the high values, a direct relation was observed. In addition, it was found that the effect of surfactant concentration on particle size in the nanosuspension of stable iodine (¹²⁷I) was depended on the solvent temperature.

Graphical Abstract

Open in a separate windowNanoprecipitation process of stable iodine (127I) and optimization of particle size using ANNs modeling.KEY WORDS: ANNs, microfluidic, nanoprecipitation, particle size, stable iodine     

Optimization of Nanoemulsion Fabrication Using Microfluidization: Role of Surfactant Concentration on Formation and Stability

Nanoemulsions have some important potential advantages over conventional emulsions for certain commercial applications due to their optical clarity, high physical stability, and ability to increase the bioavailability of lipophilic bioactives. In this study, the factors influencing droplet size and stability in nanoemulsions fabricated from a hydrocarbon oil and an anionic surfactant were examined. Octadecane oil-in-water nanoemulsions were produced by a high pressure homogenizer (microfluidizer) using sodium dodecyl sulfate (SDS) as a model anionic surfactant. The influence of homogenization pressure, number of passes, and surfactant concentration was examined. The droplet size decreased with increasing homogenization pressure, number of passes, and surfactant concentration. Nanoemulsions with low turbidity and small droplet diameters (≈62 nm) could be produced under optimized conditions. Interestingly, nanoemulsions containing relatively high surfactant levels were highly susceptible to creaming when they were only passed through the homogenizer a few times, which was attributed to depletion flocculation. These results show the importance of optimizing surfactant levels to produce small droplets that are also stable to creaming.     

Applications of neural networks to recovery of biological products

Artificial neural networks (ANN) are being applied to recovery of products from fermentation broths. Recovery methods for which mathematical models are complex or non-existent are particularly suitable for control and analysis by ANNs. Use and potential of artificial neural networks for product recovery applications are reviewed.     

Evaluation of soil aggregate stability in Algerian northwestern soils using pedotransfer functions and artificial neural networks

Since, aggregate stability is the main physical property regulating erodibility; its observations can act as a useful indicator for monitoring and managing soil degradation. In this context, this study carried out in the alluvial plain of Cheliff, a semi-arid area aimed to predict aggregate stability through Mean Weight Diameter (MWD), using pedotransfer functions (PTFs) with different stratifications (textural, salinity and organic-textural) and artificial neural networks (ANNs). Results showed that the best MWD predictions were those related to organic-textural PTFs, in this stratification the silty-clay moderately rich OM class showed the highest significant determination coefficient R² (0.65) and the lowest mean square error (0.03), whereas, the textural and salinity PTFs were a very weak predictors with a very low R². It was also found that the performances of ANNs in predicting MWD were better than those of PTFs, regarding ANNs input variables the best predictions were those obtained with a large number of input variables, furthermore, by using a large number of hidden neurons, the performances of Radial Basis Function (RBF) were better than those of Multilayer Perceptron (MLP). It was also noted that the best RBF results were always related to the Gaussian hidden activation, whereas, MLP was not related to a specific hidden activation.     

Erucic acid production using porcine pancreas lipase: Enhancement by mixed surfactants

Application of mixed surfactants coupled with statistical optimization in lipase catalyzed oil hydrolysis is presented for the first time in this study. Selective hydrolysis of brown mustard oil to erucic acid by porcine pancreas lipase was enhanced by mixed surfactants comprising of an oil-soluble nonionic surfactant (Span 80) and a watersoluble nonionic surfactant (Tween 80). The production of erucic acid was maximized using statistically designed experiments and subsequent analysis of their result by response surface methodology. The most significant variables were enzyme concentration and concentration of Tween 80. Small changes in pH and concentration of Span 80 also produced a significant change in the production of erucic acid. Temperature and speed of agitation were insignificant variables and were fixed at 35oC and 900 rpm, respectively. Under these conditions, the optimal combination of other variables were pH 9.65, 2.13 mg/g enzyme in oil, 9.8 × 10⁻³ M Span 80 (in oil), and 4 × 10⁻³ M Tween 80 (in buffer). These conditions led to formation of 99.69% of the total erucic acid in 1.25 h. Interaction of enzyme concentration with pH significantly affected erucic acid production.     

Predicting local fish species richness in the garonne river basin

The aim of this work was to predict local fish species richness in the Garonne river basin using three environmental variables (distance from the source, elevation and catchment area J. Commonly, patterns of fish species richness have been investigated using simple or multi-linear statistical models. Here, we used backpropagation of artificial neural networks (ANNs) to develop stochastic models of local fish diversity. Two independent data collections were used, the first one to build and test the model; the second one to validate the model. Correlation coefficients between observed values and predicted values both in the testing and the validation procedures were highly significant (r = 0.904, P< 0.001 and r = 0.822, P< 0.001, respectively J. The ANN model obtained using only three environmental variables succeeded in explaining ca 70 % of the total variation in local fish species richness. Through these findings, ANNs can be seen as a powerful predictive tool compared to traditional modelling approaches.     

The use of artificial neural networks for Aegilops species identification based on analysis of D genomes

The possibilities of the use of artificial neural networks (ANNs) for identification of some polyploid species of genus Aegilops based on the idiograms of their D genomes were demonstrated.     

The Use of Artificial Neural Networks for Species Identification of Aegilops Based on Analysis of D Genomes

The possibilities of the use of artificial neural networks (ANNs) for identification of some polyploid species of genus Aegilopsbased on the idiograms of theirDgenomes were demonstrated.     

A simulation study comparing the impact of experimental error on the performance of experimental designs and artificial neural networks used for process screening

Many variables and their interactions can affect a biotechnological process. Testing a large number of variables and all their possible interactions is a cumbersome task and its cost can be prohibitive. Several screening strategies, with a relatively low number of experiments, can be used to find which variables have the largest impact on the process and estimate the magnitude of their effect. One approach for process screening is the use of experimental designs, among which fractional factorial and Plackett–Burman designs are frequent choices. Other screening strategies involve the use of artificial neural networks (ANNs). The advantage of ANNs is that they have fewer assumptions than experimental designs, but they render black-box models (i.e., little information can be extracted about the process mechanics). In this paper, we simulate a biotechnological process (fed-batch growth of bakers yeast) to analyze and compare the effect of random experimental errors of different magnitudes and statistical distributions on experimental designs and ANNs. Except for the situation in which the error has a normal distribution and the standard deviation is constant, it was not possible to determine a clear-cut rule for favoring one screening strategy over the other. Instead, we found that the data can be better analyzed using both strategies simultaneously.     

The Use of Artificial Neural Networks for Optimizing Polydispersity Index (PDI) in Nanoprecipitation Process of Acetaminophen in Microfluidic Devices

Artificial neural networks (ANNs) were used in this study to determine factors that control the polydispersity index (PDI) in an acetaminophen nanosuspension which was prepared using nanoprecipitation in microfluidic devices. The PDI of prepared formulations was measured by dynamic light scattering. Afterwards, the ANNs were applied to model the data. Four independent variables, namely, surfactant concentration, solvent temperature, and flow rate of solvent and antisolvent were considered as input variables, and the PDI of acetaminophen nanosuspension was taken as the output variable. The response surfaces, generated as 3D graphs after modeling, were used to survey the interactions happening between the input variables and the output variable. Comparison of the response surfaces indicated that the antisolvent flow rate and the solvent temperature have reverse effect on the PDI, whereas solvent flow rate has direct relation with PDI. Also, the effect of the concentration of the surfactant on the PDI was found to be indirect and less influential. Overall, it was found that minimum PDI may be obtained at high values of antisolvent flow rate and solvent temperature, while the solvent flow rate should be kept to a minimum.     

Use of different approaches to model presence/absence of Salmo marmoratus in Piedmont (Northwestern Italy)

In Piedmont (Italy) the environmental changes due to human impact have had profound effects on rivers and their inhabitants. Thus, it is necessary to develop practical tools providing accurate ecological assessments of river and species conditions. We focus our attention on Salmo marmoratus, an endangered salmonid which is characteristic of the Po river system in Italy. In order to contribute to the management of the species, four different approaches were used to assess its presence: discriminant function analysis, logistic regression, decision tree models and artificial neural networks. Either all the 20 environmental variables measured in the field or the 7 coming from feature selection were used to classify sites as positive or negative for S. marmoratus. The performances of the different models were compared. Discriminant function analysis, logistic regression, and decision tree models (unpruned and pruned) had relatively high percentages of correctly classified instances. Although neither tree-pruning technique improved the reliability of the models significantly, they did reduce the tree complexity and hence increased the clarity of the models. The artificial neural network (ANN) approach, especially the model built with the 7 inputs coming from feature selection, showed better performance than all the others. The relative contribution of each independent variable to this model was determined by using the sensitivity analysis technique. Our findings proved that the ANNs were more effective than the other classification techniques. Moreover, ANNs achieved their high potentials when they were applied in models used to make decisions regarding river and conservation management.     

Importance of feature selection in decision-tree and artificial-neural-network ecological applications. Alburnus alburnus alborella: A practical example

Recent advances in computing technology have increased interest in applying data mining to ecology. Machine learning is one of the methods used in most of these data mining applications. As is well known, approximately 80% of the resources in most data mining applications are devoted to cleaning and preprocessing the data. However, there are few studies on preprocessing the ecological data used as the input in these data mining systems. In this study, we use four different feature selection methods (χ², Information Gain, Gain Ratio, and Symmetrical Uncertainty) and evaluate their effectiveness in preprocessing the input data to be used for inducing artificial neural networks (ANNs) and decision trees (DTs). The presence/absence of fish is the data item used to illustrate our models. Feature selection is fundamental in order to increase the performances of the models obtained. Accuracy of classification improves when a small set of optimally selected features is used. DTs and ANNs are very useful tools when applied to modeling presence/absence of Alburnus alburnus alborella. ANNs generally performed better than DT models.     

Use of artificial neural networks for analysis of the factors affecting particle size in mebudipine nanoemulsion

In this study, a nanoemulsion containing mebudipine [composed of ethyl oleate (oil phase), Tween 80 (T80), Span 80 (S80) (surfactants), polyethylene glycol 400, ethanol (cosurfactants), and deionized water] was prepared with the aim of improving its bioavailability for an effective antihypertensive therapy. Particle size of the formulation was measured by dynamic light scattering. Then, artificial neural networks were used in identifying factors that influence the particle size of the nanoemulsion. Three variables, namely, amount of surfactant system (T80?+?S80), amount of polyethylene glycol, and amount of ethanol as cosurfactants, were considered as input values and the particle size was used as output. The developed model showed that all the three inputs had some degrees of effect on particles size: increasing the value of each input decreased the size. Furthermore, amount of surfactant was found to be the dominant factor in controlling the final particle size of nanoemulsion.

Communicated by Ramaswamy H. Sarma     

Training radial basis function neural networks: effects of training set size and imbalanced training sets

Obtaining training data for constructing artificial neural networks (ANNs) to identify microbiological taxa is not always easy. Often, only small data sets with different numbers of observations per taxon are available. Here, the effect of both size of the training data set and of an imbalanced number of training patterns for different taxa is investigated using radial basis function ANNs to identify up to 60 species of marine microalgae. The best networks trained to discriminate 20, 40 and 60 species respectively gave overall percentage correct identification of 92, 84 and 77%. From 100 to 200 patterns per species was sufficient in networks trained to discriminate 20, 40 or 60 species. For 40 and 60 species data sets an imbalance in the number of training patterns per species always affected training success, the greater the imbalance the greater the effect. However, this could be largely compensated for by adjusting the networks using a posteriori probabilities, estimated as network output values.     

Rapid identification using pyrolysis mass spectrometry and artificial neural networks of Propionibacterium acnes isolated from dogs

Curie-point pyrolysis mass spectra were obtained from reference Propionibacterium strains and canine isolates. Artificial neural networks (ANNs) were trained by supervised learning (with the back-propagation algorithm) to recognize these strains from their pyrolysis mass spectra; all the strains isolated from dogs were identified as human wild type P. acnes. This is an important nosological discovery, and demonstrates that the combination of pyrolysis mass spectrometry and ANNs provides an objective, rapid and accurate identification technique. Bacteria isolated from different biopsy specimens from the same dog were found to be separate strains of P. acnes , demonstrating a within-animal variation in microflora. The classification of the canine isolates by Kohonen artificial neural networks (KANNs) was compared with the classical multivariate techniques of canonical variates analysis and hierarchical cluster analysis, and found to give similar results. This is the first demonstration, within microbiology, of KANNs as an unsupervised clustering technique which has the potential to group pyrolysis mass spectra both automatically and relatively objectively.     

Prediction of problematic wine fermentations using artificial neural networks

Artificial neural networks (ANNs) have been used for the recognition of non-linear patterns, a characteristic of bioprocesses like wine production. In this work, ANNs were tested to predict problems of wine fermentation. A database of about 20,000 data from industrial fermentations of Cabernet Sauvignon and 33 variables was used. Two different ways of inputting data into the model were studied, by points and by fermentation. Additionally, different sub-cases were studied by varying the predictor variables (total sugar, alcohol, glycerol, density, organic acids and nitrogen compounds) and the time of fermentation (72, 96 and 256 h). The input of data by fermentations gave better results than the input of data by points. In fact, it was possible to predict 100% of normal and problematic fermentations using three predictor variables: sugars, density and alcohol at 72 h (3 days). Overall, ANNs were capable of obtaining 80% of prediction using only one predictor variable at 72 h; however, it is recommended to add more fermentations to confirm this promising result.     

A Cremophor-Free Self-Microemulsified Delivery System for Intravenous Injection of Teniposide: Evaluation In Vitro and In Vivo

In order to tackle the problems on low water solubility of teniposide, involvement of toxic surfactant in its injection, and the poor stability during infusion, a Cremophor-free teniposide self-microemulsified drug delivery system (TEN-SMEDDS) was prepared for the first time, characterized, and evaluated in comparison with teniposide injection (VUMON) in vitro and in vivo. The optimized formulation contained N, N-dimethylacetamide, medium-chain triglyceride, lecithin, and dehydrated alcohol besides teniposide. The TEN-SMEDDS could form fine droplets with mean diameter of 282 ± 21 nm and zeta potential of −7.5 ± 1.7 mV after dilution with 5% glucose, which were stable within 4 h. The release of teniposide from TEN-SMEDDS and VUMON was similar. However, the pharmacokinetic behavior of TEN-SMEDDS in rats was different from that of VUMON, evidenced by the lower area under the concentration–time curve and larger volume of distribution in emulsion group. Finally, TEN-SMEDDS was found to distribute more teniposide in most tissues, especially in reticuloendothelial system, after intravenous administration to rats. Importantly, brain drug level in TEN-SMEDDS group was higher than or similar to that in control group, although the emulsion system had a lower plasma drug concentration. In conclusion, the novel SMEDDS prepared here, without toxic surfactant and as an oil solution before use, may be potential for clinical use due to its low toxicity and high store stability. It may be favorable for the treatment of some tumors like cerebroma, since it may achieve the relatively higher drug level in brain but lower blood concentration.KEY WORDS: characterization, pharmacokinetics, self-microemulsified drug delivery system, teniposide, tissue distribution     

Evaluation of Tadalafil Nanosuspensions and Their PEG Solid Dispersion Matrices for Enhancing Its Dissolution Properties

The aim of this work was to prepare and evaluate Tadalafil nanosuspensions and their PEG 4000 solid dispersion matrices to enhance its dissolution rate. Nanosuspensions were prepared by precipitation/ultrasonication technique at 5°C where different stabilizers were screened for stabilization. Nanosuspensions were characterized in terms of particle size and charge. Screening process limited suitable stabilizers into structurally related surfactants composed of a mixture of Tween80 and Span80 at 1:1 ratio (in percent, weight/volume) in adjusted alkaline pH (named TDTSp-OH). The surfactant mixture aided the production of nanosuspensions with an average particle size of 193 ± 8 nm and with short-term stability sufficient for further processing. Solid dispersion matrices made of dried Tadalafil nanosuspensions or dried Tadalafil raw powder suspensions and PEG 4000 as a carrier were prepared by direct compression. Drying was performed via dry heat or via freeze dry. Drug release studies showed that, in general, tablet formulations made of freeze-dried product exhibited faster initial release rates than the corresponding tablets made of oven-dried products which could be attributed to possible larger crystal growth and larger crushing strengths of oven-dried formulations. At best, 60% of drug was released from solid dispersion matrices, while more than 90% of drug was released from TDTSp-OH nanosuspension within the first 5 min. In conclusion, Tadalafil nanosuspensions obtained using a mixed surfactant system provided rapid dissolution rates of Tadalafil that can theoretically enhance its bioavailability.KEY WORDS: nanosuspension, particle size, solid dispersion, stabilizer, tablets, Tadalafil     

Investigation of Formulation Variables and Excipient Interaction on the Production of Niosomes

The aim of this study was to investigate the effects of formulation and process variables on the properties of niosomes formed from Span 40 as nonionic surfactant. A variety of formulations encapsulating Paclitaxel, a hydrophobic model drug, were prepared using different dicetyl phosphate (DCP) and Span 40-cholesterol (1:1) amounts. Formulations were optimized by multiple regression analysis to evaluate the changes on niosome characteristics such as entrapment efficiency, particle size, polydispersity index, zeta potential and in vitro drug release. Multiple regression analysis revealed that as Span 40-cholesterol amounts in the formulations were increased, zeta potential and percent of drug released at 24th hour were decreased. Besides, DCP was found to be effective on increasing niosome size. As a process variable, the effect of sonication was observed and findings revealed an irreversible size reduction on Span 40 niosomes after probe sonication. Monodisperse small sized (133 ± 6.01 nm) Span 40 niosomes entrapping 98.2% of Paclitaxel with a weight percentage of 3.64% were successfully prepared. The drug–excipient interactions in niosomes were observed by differential scanning calorimetry and X-ray powder diffraction analysis. Both techniques suggest the conversion of PCTs’ crystal structure to amorphous form. The thermal analyses demonstrate the high interaction between drug and surfactant that explains high entrapment efficiency. After 3-month storage, niosomes preserved their stability in terms of drug amount and particle size. Overall, this study showed that Span 40 niosomes with desired properties can be prepared by changing the content and production variables.Key words: drug delivery systems, drug release, multiple regression, niosomes, paclitaxel