Artificial neural network as a novel method to optimize pharmaceutical formulations

Could artificial intelligence revolutionize the development of nanovectors for gene therapy and mRNA vaccines?

Gene therapy enables the introduction of nucleic acids like DNA and RNA into host cells, and is expected to revolutionize the treatment of a wide range of diseases. This growth has been further accelerated by the discovery of CRISPR/Cas technology, which allows accurate genomic editing in a broad range of cells and organisms in vitro and in vivo. Despite many advances in gene delivery and the development of various viral and non-viral gene delivery vectors, the lack of highly efficient non-viral systems with low cellular toxicity remains a challenge. The application of cutting-edge technologies such as artificial intelligence (AI) has great potential to find new paradigms to solve this issue. Herein, we review AI and its major subfields including machine learning (ML), neural networks (NNs), expert systems, deep learning (DL), computer vision and robotics. We discuss the potential of AI-based models and algorithms in the design of targeted gene delivery vehicles capable of crossing extracellular and intracellular barriers by viral mimicry strategies. We finally discuss the role of AI in improving the function of CRISPR/Cas systems, developing novel nanobots, and mRNA vaccine carriers.

Microfluidic Manufacturing of Liposomes: Development and Optimization by Design of Experiment and Machine Learning

Liposomes constitute the most exploited drug-nanocarrier with several liposomal drugs on the market. Microfluidic-based preparation methods stand up as a promising approach with high reproducibility and the ability to scale up. In this study, liposomes composed of DOPC, cholesterol, and DSPE-PEG 2000 with different molar ratios were fabricated using a microfluidic system. Process and conditions were optimized by applying design of experiments (DoE) principles. Furthermore, data were used to build an artificial neural network (ANN) model, to predict size and polydispersity index (PDI). Sets of runs were designed by DoE and performed on a micromixer microfluidic chip. Lipids' molar ratio and the process parameters, i.e. total flow rate (TFR) and flow rate ratio (FRR), were found to be the most influential factors on the formation of vesicles with target size and PDI under 100 nm and lower than 0.2, respectively. Size and PDI were predicted by the ANN model for 3 preparations with defined experimental conditions. The results showed no significant difference in size and PDI between the preparations and their values calculated with the ANN. In conclusion, production of optimized liposomes with high reproducibility was achieved by the application of microfluidic manufacturing processes, DoE, and Artificial Intelligence (AI). Microfluidic-based preparation methods assisted by computational tools would enable a faster development and clinical transfer of nanobased medications.

Computational pharmaceutics - A new paradigm of drug delivery

In recent decades pharmaceutics and drug delivery have become increasingly critical in the pharmaceutical industry due to longer time, higher cost, and less productivity of new molecular entities (NMEs). However, current formulation development still relies on traditional trial-and-error experiments, which are time-consuming, costly, and unpredictable. With the exponential growth of computing capability and algorithms, in recent ten years, a new discipline named "computational pharmaceutics" integrates with big data, artificial intelligence, and multi-scale modeling techniques into pharmaceutics, which offered great potential to shift the paradigm of drug delivery. Computational pharmaceutics can provide multi-scale lenses to pharmaceutical scientists, revealing physical, chemical, mathematical, and data-driven details ranging across pre-formulation studies, formulation screening, in vivo prediction in the human body, and precision medicine in the clinic. The present paper provides a comprehensive and detailed review in all areas of computational pharmaceutics and "Pharma 4.0", including artificial intelligence and machine learning algorithms, molecular modeling, mathematical modeling, process simulation, and physiologically based pharmacokinetic (PBPK) modeling. We not only summarized the theories and progress of these technologies but also discussed the regulatory requirements, current challenges, and future perspectives in the area, such as talent training and a culture change in the future pharmaceutical industry.

Mathematical Modelling of Biosensing Platforms Applied for Environmental Monitoring

In recent years, mathematical modelling has known an overwhelming integration in different scientific fields. In general, modelling is used to obtain new insights and achieve more quantitative and qualitative information about systems by programming language, manipulating matrices, creating algorithms and tracing functions and data. Researchers have been inspired by these techniques to explore several methods to solve many problems with high precision. In this direction, simulation and modelling have been employed for the development of sensitive and selective detection tools in different fields including environmental control. Emerging pollutants such as pesticides, heavy metals and pharmaceuticals are contaminating water resources, thus threatening wildlife. As a consequence, various biosensors using modelling have been reported in the literature for efficient environmental monitoring. In this review paper, the recent biosensors inspired by modelling and applied for environmental monitoring will be overviewed. Moreover, the level of success and the analytical performances of each modelling-biosensor will be discussed. Finally, current challenges in this field will be highlighted.

Application of machine learning to predict monomer retention of therapeutic proteins after long term storage

An important aspect of initial developability assessments as well formulation development and selection of therapeutic proteins is the evaluation of data obtained under accelerated stress condition, i.e. at elevated temperatures. We propose the application of artificial neural networks (ANNs) to predict long term stability in real storage condition from accelerated stability studies and other high-throughput biophysical properties e.g. the first apparent temperature of unfolding (T_m). Our models have been trained on therapeutic relevant proteins, including monoclonal antibodies, in various pharmaceutically relevant formulations. Further, we developed network architectures with good prediction power using the least amount of input features, i.e. experimental effort to train the network. This provides an empiric means to highlight the most important parameters in the prediction of real-time protein stability. Further, several models were developed by a different validation means (i.e. leave-one-protein-out cross-validation) to test the robustness and the limitations of our approach. Finally, we apply surrogate machine learning algorithms (e.g. linear regression) to build trust in the ANNs decision making procedure and to highlight the connection between the leading inputs and the outputs.

The significance of artificial intelligence in drug delivery system design

Over the last decade, increasing interest has been attracted towards the application of artificial intelligence (AI) technology for analyzing and interpreting the biological or genetic information, accelerated drug discovery, and identification of the selective small-molecule modulators or rare molecules and prediction of their behavior. Application of the automated workflows and databases for rapid analysis of the huge amounts of data and artificial neural networks (ANNs) for development of the novel hypotheses and treatment strategies, prediction of disease progression, and evaluation of the pharmacological profiles of drug candidates may significantly improve treatment outcomes. Target fishing (TF) by rapid prediction or identification of the biological targets might be of great help for linking targets to the novel compounds. AI and TF methods in association with human expertise may indeed revolutionize the current theranostic strategies, meanwhile, validation approaches are necessary to overcome the potential challenges and ensure higher accuracy. In this review, the significance of AI and TF in the development of drugs and delivery systems and the potential challenging issues have been highlighted.

Mathematical Modeling of Release Kinetics from Supramolecular Drug Delivery Systems

Embedding of active substances in supramolecular systems has as the main goal to ensure the controlled release of the active ingredients. Whatever the final architecture or entrapment mechanism, modeling of release is challenging due to the moving boundary conditions and complex initial conditions. Despite huge diversity of formulations, diffusion phenomena are involved in practically all release processes. The approach in this paper starts, therefore, from mathematical methods for solving the diffusion equation in initial and boundary conditions, which are further connected with phenomenological conditions, simplified and idealized in order to lead to problems which can be analytically solved. Consequently, the release models are classified starting from the geometry of diffusion domain, initial conditions, and conditions on frontiers. Taking into account that practically all solutions of the models use the separation of variables method and integral transformation method, two specific applications of these methods are included. This paper suggests that "good modeling practice" of release kinetics consists essentially of identifying the most appropriate mathematical conditions corresponding to implied physicochemical phenomena. However, in most of the cases, models can be written but analytical solutions for these models cannot be obtained. Consequently, empiric models remain the first choice, and they receive an important place in the review.

Additive Manufactured Sandwich Composite/ABS Parts for Unmanned Aerial Vehicle Applications

Fused deposition modelling (FDM) is one of most popular 3D printing techniques of thermoplastic polymers. Nonetheless, the poor mechanical strength of FDM parts restricts the use of this technology in functional parts of many applications such as unmanned aerial vehicles (UAVs) where lightweight, high strength, and stiffness are required. In the present paper, the fabrication process of low-density acrylonitrile butadiene styrenecarbon (ABS) with carbon fibre reinforced polymer (CFRP) sandwich layers for UAV structure is proposed to improve the poor mechanical strength and elastic modulus of printed ABS. The composite sandwich structures retains FDM advantages for rapid making of complex geometries, while only requires simple post-processing steps to improve the mechanical properties. Artificial neural network (ANN) was used to investigate the influence of the core density and number of CFRP layers on the mechanical properties. The results showed an improvement of specific strength and elastic modulus with increasing the number of CFRP. The specific strength of the samples improved from 20 to 145 KN·m/kg while the Young's modulus increased from 0.63 to 10.1 GPa when laminating the samples with CFRP layers. On the other hand, the core density had no significant effect on both specific strength and elastic modulus. A case study was undertaken by applying the CFRP/ABS/CFRP sandwich structure using the proposed method to manufacture improved dual-tilting clamps of a quadcopter UAV.

The Future of Pharmaceutical Manufacturing Sciences

The entire pharmaceutical sector is in an urgent need of both innovative technological solutions and fundamental scientific work, enabling the production of highly engineered drug products. Commercial-scale manufacturing of complex drug delivery systems (DDSs) using the existing technologies is challenging. This review covers important elements of manufacturing sciences, beginning with risk management strategies and design of experiments (DoE) techniques. Experimental techniques should, where possible, be supported by computational approaches. With that regard, state-of-art mechanistic process modeling techniques are described in detail. Implementation of materials science tools paves the way to molecular-based processing of future DDSs. A snapshot of some of the existing tools is presented. Additionally, general engineering principles are discussed covering process measurement and process control solutions. Last part of the review addresses future manufacturing solutions, covering continuous processing and, specifically, hot-melt processing and printing-based technologies. Finally, challenges related to implementing these technologies as a part of future health care systems are discussed.

Variations of glutamate concentration within synaptic cleft in the presence of electromagnetic fields: an artificial neural networks study

Glutamate is an excitatory neurotransmitter that is released by the majority of central nervous system synapses and is involved in developmental processes, cognitive functions, learning and memory. Excessive elevated concentrations of Glu in synaptic cleft results in neural cell apoptosis which is called excitotoxicity causing neurodegenerative diseases. Hence, we investigated the possibility of extremely low frequency electromagnetic fields (ELF-EMF) as a risk factor which is able to change Glu concentration in synaptic clef. Synaptosomes as a model of nervous terminal were exposed to ELF-EMF for 15-55 min in flux intensity range from 0.1 to 2 mT and frequency range from 50 to 230 Hz. Finally, all raw data by INForm v4.02 software as an artificial neural network program was analyzed to predict the effect of whole mentioned range spectra. The results showed the tolerance of all effects between the ranges from -35 to +40 % compared to normal state when glutamatergic systems exposed to ELF-EMF. It indicates that glutamatergic system attempts to compensate environmental changes though release or reuptake in order to keep the system safe. Regarding to the wide range of ELF-EMF acquired in this study, the obtained outcomes have potential for developing treatments based on ELF-EMF for some neurological diseases; however, in vivo experiments on the cross linking responses between glutamatergic and cholinergic systems in the presence of ELF-EMF would be needed.

Defining the critical material attributes of lactose monohydrate in carrier based dry powder inhaler formulations using artificial neural networks

The study aimed to establish a function-based relationship between the physical and bulk properties of pre-blended mixtures of fine and coarse lactose grades with the in vitro performance of an adhesive active pharmaceutical ingredient (API). Different grades of micronised and milled lactose (Lactohale (LH) LH300, LH230, LH210 and Sorbolac 400) were pre-blended with coarse grades of lactose (LH100, LH206 and Respitose SV010) at concentrations of 2.5, 5, 10 and 20 wt.%. The bulk and rheological properties and particle size distributions were characterised. The pre-blends were formulated with micronised budesonide and in vitro performance in a Cyclohaler device tested using a next-generation impactor (NGI) at 90 l/min. Correlations between the lactose properties and in vitro performance were established using linear regression and artificial neural network (ANN) analyses. The addition of milled and micronised lactose fines with the coarse lactose had a significant influence on physical and rheological properties of the bulk lactose. Formulations of the different pre-blends with budesonide directly influenced in vitro performance attributes including fine particle fraction, mass median aerodynamic diameter and pre-separator deposition. While linear regression suggested a number of physical and bulk properties may influence in vitro performance, ANN analysis suggested the critical parameters in describing in vitro deposition patterns were the relative concentrations of lactose fines % < 4.5 μm and % < 15 μm. These data suggest that, for an adhesive API, the proportion of fine particles below % < 4.5 μm and % < 15 μm could be used in rational dry powder inhaler formulation design.

Combinatorial approach in the design of multifunctional polymeric nano-delivery systems for cancer therapy

This update summarizes the recent advances in combinatorial design of polymeric material for developing multifunctional nanovectors to deliver nucleic acids and chemodrugs for cancer therapy.

MODELING AND OPTIMIZATION OF A PHARMACEUTICAL FORMULATION SYSTEM USING RADIAL BASIS FUNCTION NETWORK

A Pharmaceutical formulation is composed of several formulation factors and process variables. Quantitative model based pharmaceutical formulation involves establishing mathematical relations between the formulation variables and the resulting responses, and optimizing the formulation conditions. In a formulation system involving several objectives, the desirable formulation conditions for one property may not always be desirable for other characteristics, thus leading to the problem of conflicting objectives. Therefore, efficient modeling and optimization techniques are needed to devise an optimal formulation system. In this work, a novel method based on radial basis function network (RBFN) is proposed for modeling and optimization of pharmaceutical formulations involving several objectives. This method has the advantage that it automatically configures the RBFN using a hierarchically self organizing learning algorithm while establishing the network parameters. This method is evaluated by using a trapidil formulation system as a test bed and compared with that of a response surface method (RSM) based on multiple regression. The simulation results demonstrate the better performance of the proposed RBFN method for modeling and optimization of pharmaceutical formulations over the regression based RSM technique.

Creation of a tablet database containing several active ingredients and prediction of their pharmaceutical characteristics based on ensemble artificial neural networks

A tablet database containing several active ingredients for a standard tablet formulation was created. Tablet tensile strength (TS) and disintegration time (DT) were measured before and after storage for 30 days at 40 degrees C and 75% relative humidity. An ensemble artificial neural network (EANN) was used to predict responses to differences in quantities of excipients and physical-chemical properties of active ingredients in tablets. Most classical neural networks involve a tedious trial and error approach, but EANNs automatically determine basal key parameters, which ensure that an optimal structure is rapidly obtained. We compared the predictive abilities of EANNs in which the following kinds of training algorithms were used: linear, radial basis function, general regression (GR), and multilayer perceptron. The GR EANN predicted pharmaceutical responses such as TS and DT most accurately, as evidenced by high correlation coefficients in a leave-some-out cross-validation procedure. When used in conjunction with a tablet database, the GR EANN is capable of identifying acceptable candidate tablet formulations.

Predictive and correlative techniques for the design, optimisation and manufacture of solid dosage forms

There is much interest in predicting the properties of pharmaceutical dosage forms from the properties of the raw materials they contain. Achieving this with reasonable accuracy would aid the faster development and manufacture of dosage forms. A variety of approaches to prediction or correlation of properties are reviewed. These approaches have variable accuracy, with no single technique yet able to provide an accurate prediction of the overall properties of the dosage form. However, there have been some successes in predicting trends within a formulation series based on the physicochemical and mechanical properties of raw materials, predicting process scale-up through mechanical characterisation of materials and predicting product characteristics by process monitoring. Advances in information technology have increased predictive capability and accuracy by facilitating the analysis of complex multivariate data, mapping formulation characteristics and capturing past knowledge and experience.

Advantages of neurofuzzy logic against conventional experimental design and statistical analysis in studying and developing direct compression formulations

This study has investigated the utility and potential advantages of an artificial intelligence technology - neurofuzzy logic - as a modeling tool to study direct compression formulations. The modeling performance was compare with traditional statistical analysis. From results it can be stated that the normalized error obtained from neurofuzzy logic was lower. Compared to the multiple regression analysis neurofuzzy logic showed higher accuracy in prediction for the five outputs studied. Rule sets generated by neurofuzzy logic are completely in agreement with the findings based on statistical analysis and advantageously generate understandable and reusable knowledge. Neurofuzzy logic is easy and rapid to apply and outcomes provided knowledge not revealed via statistical analysis.

Orthogonal projection to latent structures combined with artificial neural networks in non-destructive analysis of Ampicillin powder

A new method orthogonal projection to latent structures (O-PLS) combined with artificial neural networks is investigated for non-destructive determination of Ampicillin powder via near-infrared (NIR) spectroscopy. The modern NIR spectroscopy analysis technique is efficient, simple and non-destructive, which has been used in chemical analysis in diverse fields. Be a preprocessing method, O-PLS provides a way to remove systematic variation from an input data set X not correlated to the response set Y, and does not disturb the correlation between X and Y. In this paper, O-PLS pretreated spectral data was applied to establish the ANN model of Ampicillin powder, in this model, the concentration of Ampicillin as the active component was determined. The degree of approximation was employed as the selective criterion of the optimum network parameters. In order to compare the OPLS-ANN model, the calibration models that using first-derivative and second-derivative preprocessing spectra were also designed. Experimental results showed that the OPLS-ANN model was the best.

Prediction of Dissolution Profiles of Acetaminophen Beads Using Artificial Neural Networks

Immediate release acetaminophen (APAP) beads with 40% drug loading were prepared using the extrusion-spheronization process. Eighteen batches of beads were prepared based on a full factorial design by varying process variables such as extruder type, extruder screw speed, spheronization speed, and spheronization time. An in vitro dissolution test was carried out using the USP 27 Apparatus II (paddle) method. Artificial Neural Network (ANN) models were developed based on the aforementioned process variables and dissolution data. The trained ANN models were used to predict the dissolution profiles of APAP from the beads, which were prepared with various processing conditions. For training the ANN models, process variables were used as inputs, and percent drug released from APAP beads was used as the output. The dissolution data from one out of 18 batches of APAP beads was selected as the validation data set. The dissolution data of other 17 batches were used to train the ANN models using the ANN software (AI Trilogy) with two different training strategies, namely, neural and genetic. The validation results showed that the ANN model trained with the genetic strategy had better predictability than the one trained with the neural strategy. The ANN model trained with the genetic strategy was then used to predict the drug release profiles of two new batches of APAP beads, which were prepared with process variables that were not used during the ANN model training process. However, the process variables used to prepare the two new batches of APAP beads were within the confines of the process variables used to prepare the 18 batches. The actual drug release profile of these two batches of APAP beads was similar to the ones predicted by the trained and validated ANN model, as indicated by the high f2 values. Furthermore, the ANN model trained with genetic strategy was also used to optimize process variables to achieve the desired dissolution profiles. These batches of APAP beads were then actually prepared using the process variables predicted by the trained and validated ANN model. The dissolution results showed that the actual dissolution profiles of the APAP beads prepared from the predicted process variables were similar to the desired dissolution profiles.

Artificial Neural Networks: Comparison of Two Programs for Modeling a Process of Nanoparticle Preparation

Artificial Neural Networks (ANNs) were used to predict nanoparticle size and micropore surface area of polylactic acid nanoparticles, prepared by a double emulsion method. Different batches were prepared while varying polymer and surfactant concentration, as well as homogenization pressure. Two commercial ANNs programs were evaluated: Neuroshell Predictor, a black-box software adopting both neural and genetic strategies, and Neurosolutions, allowing a step-by-step building of the network. Results were compared to those obtained by statistical method. Predictions from ANNs were more accurate than those calculated using non-linear regression. Neuroshell Predictor allowed quantification of the relative importance of the inputs. Furthermore, by varying the network topology and parameters using Neurosolutions, it was possible to obtain output values which were closer to experimental values. Therefore, ANNs represent a promising tool for the analysis of processes involving preparation of polymeric carriers and for prediction of their physical properties.

An investigation into the usefulness of generalized regression neural network analysis in the development of level A in vitro–in vivo correlation

Quantitative correlations between in vivo and in vitro data (IVIVC) reduce the number of human in vivo studies, thus decreasing the overall time and expenses necessary for the development of optimal drug product formulation. Although linear regression analysis represents the simplest relationship, it is recognized that IVIVC should not be limited to linear relationship. With regards to the implementation of non-linear IVIVC models and the ability of artificial neural network (ANN) computing to cope with non-linear relationships, the usefulness of ANN analysis in the development of IVIVC merits further evaluation. The present paper is an attempt to develop an IVIVC for model sustained release paracetamol matrix tablet formulations employing various correlation approaches based on linear and non-linear modeling of in vitro and in vivo data. Currently accepted compendial methodology was compared with the alternative approaches, involving general mixed effects model and generalized regression neural network (GRNN) analysis, in order to evaluate their usefulness for predicting the in vivo behavior of drug products. Although based on analogous approaches, data generated by GRNN were closer to those observed in vivo, leading to the higher level of IVIVC than obtained by convolution. It can be assumed that GRNN analysis was able to generalize complex relations between the output and input parameters and could account for the differences in drug release kinetics observed under various conditions in vitro, thus offering potential as a reliable and robust estimate of drug products in vivo behavior.

Performance comparison of neural network training algorithms in modeling of bimodal drug delivery

The major aim of this study was to model the effect of two causal factors, i.e. coating weight gain and amount of pectin-chitosan in the coating solution on the in vitro release profile of theophylline for bimodal drug delivery. Artificial neural network (ANN) as a multilayer perceptron feedforward network was incorporated for developing a predictive model of the formulations. Five different training algorithms belonging to three classes: gradient descent, quasi-Newton (Levenberg-Marquardt, LM) and genetic algorithm (GA) were used to train ANN containing a single hidden layer of four nodes. The next objective of the current study was to compare the performance of aforementioned algorithms with regard to predicting ability. The ANNs were trained with those algorithms using the available experimental data as the training set. The divergence of the RMSE between the output and target values of test set was monitored and used as a criterion to stop training. Two versions of gradient descent backpropagation algorithms, i.e. incremental backpropagation (IBP) and batch backpropagation (BBP) outperformed the others. No significant differences were found between the predictive abilities of IBP and BBP, although, the convergence speed of BBP is three- to four-fold higher than IBP. Although, both gradient descent backpropagation and LM methodologies gave comparable results for the data modeling, training of ANNs with genetic algorithm was erratic. The precision of predictive ability was measured for each training algorithm and their performances were in the order of: IBP, BBP>LM>QP (quick propagation)>GA. According to BBP-ANN implementation, an increase in coating levels and a decrease in the amount of pectin-chitosan generally retarded the drug release. Moreover, the latter causal factor namely the amount of pectin-chitosan played slightly more dominant role in determination of the dissolution profiles.