Design and optimization of scored tablets with concave surface and application of Bayesian estimation for solving scaleup problem
In this study, a new design for concave tablets was optimized. Mechanical properties were assessed in experimental trials and used to simulate results for a broader range of parameter combinations. Viscovery SOMine is used to cluster the simulation results and identify parameter regions with optimal values in the 4-dimensional target space.
Effect of different direct compaction grades of mannitol on the storage stability of tablet properties investigated using a Kohonen self-organizing map and elastic net regression model
Complementary to the article "Characterization of powder- and tablet properties of different direct compaction grades of mannitol using a Kohonen self-organizing map and a Lasso regression model" by Kosugi et. al. this article analyzes the influence of Mannitol products on the storage properties of produced tablets. Again Viscovery SOMine is used to cluster a multi-dimensional target space.
Characterization of powder- and tablet properties of different direct compaction grades of mannitol using a Kohonen self-organizing map and a Lasso regression model
In this study, the effect of certain properties of different mannitol products on the tablets produced with them is analyzed. Viscovery SOMine is used to cluster the tablets with respect to multiple target variables. On top of the cluster model a Lasso regression is calculated to quantify the influence of the different properties.
A comparative study of disintegration actions of various disintegrants using Kohonen's self-organizing maps
The aim of this article is to better understand disintegration actions for pharmaceutical tablets and disintegrant powders. For this purpose, experimental data for water absorption and disintegration under different experimental conditions are analyzed using Viscovery SOMine.
Optimization of premix powders for tableting use
The composition of preliminary mixed powders for direct compression tableting is optimized in this paper. Viscovery SOMine was used in a bootstrap resampling approach to estimate a global optimum and its confidence interval, based on the experimental data.
Effect of surfactants and thickeners on the stability of mentholâdiphenhydramine cream identified by magnetic resonance imaging
The purpose of this paper is to find a stable formulation for menthol-diphenhydramine cream without concern about phase separation. The explorative analysis using Viscovery SOMine revealed that hydrophilic-lipophilic balance is an important factor for stabilizing effects of the surfactants, but, in addition, thickeners must be used to acquire the desired stability.
Membrane microdomain structures of liposomes and their contribution to the cellular uptake efficiency into HeLa cells
The relationship between membrane microdomain structure of liposomes and their cellular uptake efficiency is analyzed. Five distinct clusters for the microstructure properties are found using Viscovery SOMine.
Application of artificial neural network to investigate the effects of 5-fluorouracil on ribonucleotides and deoxyribonucleotides in HepG2 cells
This study analyzes the effects of the anti-cancer agent 5-fluorouracil on nucleotides in HepG2 cells. By comparing treated and untreated cells, the Viscovery SOMine cluster model revealed various effects on the deoxyribonucleotides.
Self-organizing map analysis for understanding comprehensive relationships between formulation variables, state of water, and the physical stability of pharmaceutical emulsions
Physical stability of pharmaceutical emulsions is analyzed in relation to surfactant content, water content and hydrophilicâlipophilic balance in the formulation. The Viscovery SOMine model of these emulsions shows four distinct clusters and allows analysis of the stable solutions.
Determination of the optimal cell-penetrating peptide sequence for intestinal insulin delivery based on molecular orbital analysis with self-organizing maps
The influence of different cell-penetrating peptides on the intestinal absorption rate of insulin is analyzed. Viscovery SOMine facilitates the discovery that molecular weight, basicity, lowest unoccupied molecular orbital energy, absolute hardness and chemical potential of the peptides have significant influences on absorption.
Self-organizing map analysis using multivariate data from theophylline tablets predicted by a thin-plate spline interpolation
Theophylline tablets are prepared based on a standard formulation with several varying design parameters. Viscovery SOMine is used to analyze the connection between design parameters and pharmaceutical responses (tensile strength and stability, disintegration time and stability).
Contribution of the physicochemical properties of active pharmaceutical ingredients to tablet properties identified by ensemble artificial neural networks and Kohonen's self-organizing maps
This paper analyzes the contribution of active pharmaceutical ingredients to tablet properties, such as hardness and disintegration time. For this purpose, an optimal base formulation for a placebo tablet is calculated. Subsequently, tablets with active pharmaceutical ingredients are prepared using this base formulation, and Viscovery SOMine is used to order the tablet physiochemical properties and analyze the distribution of influences on tablet properties.
Reliability evaluation of the design space of the granulation process of mefenamic acid tablets using a bootstrap resampling technique
Mefenamic acid tablets are optimized with respect to dissolution time, hardness and granule size. Viscovery SOMine is used for design space visualization and clustering.
A statistical approach to the development of a transdermal delivery system for ondansetron
Ondansetron hydrogel is optimized with respect to skin permeability, absence of skin irritation and pharmacological effect. Viscovery SOMine is used to visualize the optimal solutions in design space and perform a trade-off analysis.
Latent structure analysis in pharmaceutical formulations using Kohonenâs self-organizing map and a Bayesian network
In this article, the design space of hydrophilic matrix tablets containing diltiazem hydrochloride is visualized. Viscovery SOMine is used to analyze the distribution of various response variables, such as turbidity and viscosity of the polymer solution and dissolution times.
Phase behavior in a ternary lipid membrane estimated using a nonlinear response surface method and Kohonenâs self-organizing map
The phase behavior of ternary lipid membranes with different lipid compositions is analyzed. Viscovery SOMine is used to cluster the prepared membranes with respect to observed anisotropy at different temperatures.
Role of individual test samples in optimal solutions in pharmaceuticals predicted using a nonlinear response surface method
This paper describes the optimization of photo-crosslinked polyacrylic acid hydrogel, using non-linear response surface method, bootstrap sampling and self-organizing-map visualization. Viscovery SOMine is used to find clusters of similar solutions and compare the local minima and their direct neighborhood.
Reliability assessment for the optimal formulations of pharmaceutical products predicted by a nonlinear response surface method
The properties of pharmaceutical ointments, emulsions and dispersions are optimized. Viscovery SOMine is used to visualize design space and validate the optimal solutions generated by a nonlinear response surface method with bootstrap resampling.
Evaluation of the reliability of nonlinear optimal solutions in pharmaceuticals using a bootstrap resampling technique in combination with Kohonenâs self-organizing maps
A reliable design technique for multi-objective optimization of pharmaceutical products, consisting of response surface method, bootstrap sampling and self-organizing maps is introduced. As a proof of concept, theophylline tablets are optimized with respect to hardness and drug-release time. Viscovery SOMine is used to analyze the locally and globally optimal solutions to find consistent candidates for design decisions.
Phytochemical informatics of traditional Chinese medicine and therapeutic relevance
In this study, 8411 compounds from 240 Chinese herbs are clustered by Viscovery SOMine with respect to modern phytochemical classifications. The resulting clusters are compared to traditional Chinese medicine classification, revealing connections between pharmacological properties and the traditional language of Chinese medicine.
Sensor combination and chemometric modelling for improved process monitoring in recombinant E. coli fed-batch cultivations
The key objective for the optimization of recombinant protein production in bacteria is to maximize the exploitation of the host cell's synthesis potential. Since there are no reliable online-sensors for key variables, it is necessary to relate available online signals to process variables using mathematical models. To improve chemometric modeling of process variables, dielectric spectroscopy and a multi-wavelength online fluorescence sensor for two-dimensional fluorescence spectroscopy are applied to a series of recombinant Escherichia coli fed-batch cultivations, applying two different process operation states. Chemometric modeling of key process variables with two different modeling techniques showed that this sensor combination greatly improved the estimation (i.e., reduce error magnitude) of process variables in recombinant E. coli cultivations, thereby enhancing process monitoring capabilities.
Optimizing recombinant microbial fermentation processes: an integrated approach
A new strategy for controlling recombinant gene expression for improving efficiency, maximizing host-vector exploitation, reducing costs, improving product consistency and accelerating product development is described. Viscovery SOMine is used to create a model for plasmid copy-number estimates. The integrated approach is useful for producing recombinant proteins on an industrial scale and for designing experiments for targeted process optimization.
Optimized data exploration of recombinant fermentations using neural network simulations
A combination of self-organizing maps and radial basis function networks is presented as a powerful tool applied to fermentation data, enabling rapid recognition of interdependencies and subsequent modeling. Viscovery SOMine is used to model the appearance and concentration of signal molecules to get a better understanding of the relations between metabolic load and recombinant protein production and improved use of the cellâs synthetic capacity.