Application of Artificial Neural Networks (ANN) in the Development of Solid Dosage Forms

ABSTRACT

The application of ANN in pharmaceutical development has been assessed using theoretical as well as typical pharmaceutical technology examples. The aim was to quantitatively describe the achieved data fitting and predicting abilities of the models developed with a view to using ANN in the development of solid dosage forms. The comparison between the ANN and a traditional statistical (i.e., response surface methodology, RSM) modeling technique was carried out using the squared correlation coefficient R². Using a highly nonlinear arbitrary function the ANN models showed better fitting (R² = 0.931 vs. R² = 0.424) as well as predicting (R² = 0.810 vs. R² = 0.547) abilities. Experimental data from a tablet compression study were fitted using two types of ANN models (i.e., multilayer perceptrons and a hybrid network composed of a self-organising feature map joined to a multilayer perceptron). The achiedved data fitting was comparable for the three methods (MLP R² = 0.911, SOFM-MLPR² = 0.850, and RSM R² = 0.897). ANN methodology represents a promising modeling technique when applied to pharmaceutical technology data sets.     

人工神经网络辅助设计褪黑素缓释片处方

应用人工神经网络模型辅助设计褪黑素缓释片处方,将HPMC粘度、HPMC、MCC 和乳糖的量作为输入变量,累积释放百分率作为输出变量,选择反向传播网络,隐含层为1层,隐含层神经元个数为6,建立人工神经网络模型,预测和评价褪黑素缓释片体外释放度,研究缓释片的释放机理.结果显示,该人工神经网络模型能很好地预测褪黑素缓释片的释放量,成功优化褪黑素缓释片处方,其释放机理为溶蚀与扩散的结合,辅料的种类和量会对药物的释放机理产生不同影响.     

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 regularisation 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 regularisation. 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.     

人工神经网络预测别嘌醇HP MC缓释片药物释放

目的用人工神经网络模型定量的预测HPMC的量和其固有黏度对药物释放的影响。方法以难溶性药物别嘌醇为模型药物,固定其他因素,HPMC的量和HPMC的固有黏度作为自变量,设计了18个处方并进行释放度检查;其中的13个处方作为训练处方,其他5个处方为验证处方,将上述的变量作为人工神经的输入,以药物在各个取样时间点的释放为输出,采用剔除一点交叉验证法建立人工神经网络模型。通过线性回归和相似因子说明人工神经网络的预测能力。结果训练和验证处方人工神经网络预测值与实际测定相符。结论建立BP人工神经网络,根据HPMC的量和其固有黏度可以定量的预测药物在各个时间点的药物释放。     

Application of Artificial Neural Networks in the Design and Optimization of a Nanoparticulate Fingolimod Delivery System Based on Biodegradable Poly(3-Hydroxybutyrate-Co-3-Hydroxyvalerate)

Formulation of a nanoparticulate Fingolimod delivery system based on biodegradable poly(3-hydroxybutyrate-co-3-hydroxyvalerate) was optimized according to artificial neural networks (ANNs). Concentration of poly(3-hydroxybutyrate-co-3-hydroxyvalerate), PVA and amount of Fingolimod is considered as the input value, and the particle size, polydispersity index, loading capacity, and entrapment efficacy as output data in experimental design study. In vitro release study was carried out for best formulation according to statistical analysis. ANNs are employed to generate the best model to determine the relationships between various values. In order to specify the model with the best accuracy and proficiency for the in vitro release, a multilayer percepteron with different training algorithm has been examined. Three training model formulations including Levenberg-Marquardt (LM), gradient descent, and Bayesian regularization were employed for training the ANN models. It is demonstrated that the predictive ability of each training algorithm is in the order of LM > gradient descent > Bayesian regularization. Also, optimum formulation was achieved by LM training function with 15 hidden layers and 20 neurons. The transfer function of the hidden layer for this formulation and the output layer were tansig and purlin, respectively. Also, the optimization process was developed by minimizing the error among the predicted and observed values of training algorithm (about 0.0341).     

药物研发过程中药学信息的利用

目的:方便药学科研人员获取药学信息,提高药物研发的效率。方法:在分析药物研发流程的基础上,按项目调研数据库、临床前研究数据库、临床研究数据库、药物评审信息数据库等4方面对国内、外药学数据库资源进行系统调研。结果与结论:项目调研数据库主要包括Pharma project数据库等事实型数据库,临床前研究数据库主要包括美国生物医学文献数据库等文摘型数据库,临床研究数据库主要包括Clinical Trial.gov数据库等,药物评审信息数据库主要包括欧洲药品审评管理局等信息数据库。如果药学科研工作站掌握了相关的药学数据库资源,就能快速准确地检索到需要的信息,从而提高药物研发的效率。     

Modeling of Semicontinuous Fluid Bed Drying of Pharmaceutical Granules With Respect to Granule Size

In the transition of the pharmaceutical industry from batchwise to continuous drug product manufacturing, the drying process has proven challenging to control and understand. In a semicontinuous fluid bed dryer, part of the ConsiGma? wet granulation line, the aforementioned production methods converge. Previous research has shown that the evolution of moisture content of the material in this system shows strong variation in function of the granule size, making the accurate prediction of this pharmaceutical critical quality attribute a complex case. In this work, the evolution of moisture content of the material in the system is modeled by a bottom-up approach. A single granule drying kinetics model is used to predict the moisture content evolution of a batch of material of a heterogeneous particle size, where it is the first time that the single granule drying mechanism is validated for different granule sizes. The batch approach was validated when the continuous material inflow rate and filling time of the dryer cell are constant. The original single granule drying kinetics model has been extended to capture the granules’ equilibrium moisture content. Finally, the influence of drying air temperature is captured well with a droplet energy balance for the granules.     

Use of Machine Learning and Artificial Intelligence to predict SARS-CoV-2 infection from Full Blood Counts in a population

Since December 2019 the novel coronavirus SARS-CoV-2 has been identified as the cause of the pandemic COVID-19. Early symptoms overlap with other common conditions such as common cold and Influenza, making early screening and diagnosis are crucial goals for health practitioners. The aim of the study was to use machine learning (ML), an artificial neural network (ANN) and a simple statistical test to identify SARS-CoV-2 positive patients from full blood counts without knowledge of symptoms or history of the individuals. The dataset included in the analysis and training contains anonymized full blood counts results from patients seen at the Hospital Israelita Albert Einstein, at São Paulo, Brazil, and who had samples collected to perform the SARS-CoV-2 rt-PCR test during a visit to the hospital. Patient data was anonymised by the hospital, clinical data was standardized to have a mean of zero and a unit standard deviation. This data was made public with the aim to allow researchers to develop ways to enable the hospital to rapidly predict and potentially identify SARS-CoV-2 positive patients.We find that with full blood counts random forest, shallow learning and a flexible ANN model predict SARS-CoV-2 patients with high accuracy between populations on regular wards (AUC = 94–95%) and those not admitted to hospital or in the community (AUC = 80–86%). Here, AUC is the Area Under the receiver operating characteristics Curve and a measure for model performance. Moreover, a simple linear combination of 4 blood counts can be used to have an AUC of 85% for patients within the community. The normalised data of different blood parameters from SARS-CoV-2 positive patients exhibit a decrease in platelets, leukocytes, eosinophils, basophils and lymphocytes, and an increase in monocytes.SARS-CoV-2 positive patients exhibit a characteristic immune response profile pattern and changes in different parameters measured in the full blood count that are detected from simple and rapid blood tests. While symptoms at an early stage of infection are known to overlap with other common conditions, parameters of the full blood counts can be analysed to distinguish the viral type at an earlier stage than current rt-PCR tests for SARS-CoV-2 allow at present. This new methodology has potential to greatly improve initial screening for patients where PCR based diagnostic tools are limited.     

Prediction of Pharmacokinetic Parameters and the Assessment of Their Variability in Bioequivalence Studies by Artificial Neural Networks

Purpose. The methodology of predicting the pharmacokinetic parameters (AUC, c_max, t_max) and the assessment of their variability in bioequivalence studies has been developed with the use of artificial neural networks. Methods. The data sets included results of 3 distinct bioequivalence studies of oral verapamil products, involving a total of 98 subjects and 312 drug applications. The modeling process involved building feedforward/backpropagation neural networks. Models for pharmacokinetic parameter prediction were also used for the assessment of their variability and for detecting the most influential variables for selected pharmacokinetic parameters. Variables of input neurons based on logistic parameters of the bioequivalence study, clinical-biochemical parameters, and the physical examination of individuals. Results. The average absolute prediction errors of the neural networks for AUC, c_max, and t_max prediction were: 30.54%, 39.56% and 30.74%, respectively. A sensitivity analysis demonstrated that for verapamil the three most influential variables assigned to input neurons were: total protein concentration, aspartate aminotransferase (AST) levels, and heart-rate for AUC, AST levels, total proteins and alanine aminotransferase (ALT) levels, for c_max, and the presence of food, blood pressure, and body-frame for t_max. Conclusions. The developed methodology could supply inclusion or exclusion criteria for subjects to be included in bioequivalence studies.     

Artificial Neural Networks: An Overview and their Use in the Analysis of the AMPHORA-3 Dataset

The Artificial Adaptive Systems (AAS) are theories with which generative algebras are able to create artificial models simulating natural phenomenon. Artificial Neural Networks (ANNs) are the more diffused and best-known learning system models in the AAS. This article describes an overview of ANNs, noting its advantages and limitations for analyzing dynamic, complex, non-linear, multidimensional processes. An example of a specific ANN application to alcohol consumption in Spain, as part of the EU AMPHORA-3 project, during 1961–2006 is presented. Study's limitations are noted and future needed research using ANN methodologies are suggested.     

高效液相色谱指纹图谱在中药制药过程中的研究进展

目的：对高效液相色谱（ HPLC）指纹图谱技术在中药制药过程中的研究进展进行综述。方法检索近年 HPLC 指纹图谱技术文献,对其进行分析整理和归纳。结果与结论 HPLC指纹图谱技术具有整体性和稳定性的特点,能较全面反映中药制药过程的质量,已成为现代中药制药工艺质控的重要方法。     

Technology Transfer from the Science of Medicine to the Real World: The Potential Role Played by Artificial Adaptive Systems

The author describes a refiguration of medical thought that originates from nonlinear dynamics and chaos theory. The coupling of computer science and these new theoretical bases coming from complex systems mathematics allows the creation of “intelligent” agents capable of adapting themselves dynamically to problems of high complexity: the artificial neural networks (ANNs). ANNs are able to reproduce the dynamic interaction of multiple factors simultaneously, allowing the study of complexity; they can also draw conclusions on an individual basis and not as average trends. These tools can allow a more efficient technology transfer from the science of medicine to the real world, overcoming many obstacles responsible for the present translational failure. They also contribute to a new holistic vision of the human subject person, contrasting the statistical reductionism that tends to squeeze or even delete the single subject, sacrificing him to his group of belongingness. A remarkable contribution to this individual approach comes from fuzzy logic, according to which there are no sharp limits between opposite things, such as wealth and disease. This approach allows one to partially escape from the probability theory trap in situations where it is fundamental to express a judgement based on a single case and favor a novel humanism directed to the management of the patient as an individual subject person.     

Application of dynamic neural networks in the modeling of drug release from polyethylene oxide matrix tablets

The main objective of this study was to demonstrate the possible use of dynamic neural networks to model diclofenac sodium release from polyethylene oxide hydrophilic matrix tablets. High and low molecular weight polymers in the range of 0.9–5 × 10⁶ have been used as matrix forming materials and 12 different formulations were prepared for each polymer. Matrix tablets were made by direct compression method. Fractions of polymer and compression force have been selected as most influential factors on diclofenac sodium release profile. In vitro dissolution profile has been treated as time series using dynamic neural networks. Dynamic networks are expected to be advantageous in the modeling of drug release. Networks of different topologies have been constructed in order to obtain precise prediction of release profiles for test formulations. Short-term and long-term memory structures have been included in the design of network making it possible to treat dissolution profiles as time series. The ability of network to model drug release has been assessed by the determination of correlation between predicted and experimentally obtained data. Calculated difference (f₁) and similarity (f₂) factors indicate that dynamic networks are capable of accurate predictions. Dynamic neural networks were compared to most frequently used static network, multi-layered perceptron, and superiority of dynamic networks has been demonstrated. The study also demonstrated differences between the used polyethylene oxide polymers in respect to drug release and suggests explanations for the obtained results.     

A quality by design approach using artificial intelligence techniques to control the critical quality attributes of ramipril tablets manufactured by wet granulation

Quality by design (QbD) is an essential part of the modern approach to pharmaceutical quality. This study was conducted in the framework of a QbD project involving ramipril tablets. Preliminary work included identification of the critical quality attributes (CQAs) and critical process parameters (CPPs) based on the quality target product profiles (QTPPs) using the historical data and risk assessment method failure mode and effect analysis (FMEA). Compendial and in-house specifications were selected as QTPPs for ramipril tablets. CPPs that affected the product and process were used to establish an experimental design. The results thus obtained can be used to facilitate definition of the design space using tools such as design of experiments (DoE), the response surface method (RSM) and artificial neural networks (ANNs). The project was aimed at discovering hidden knowledge associated with the manufacture of ramipril tablets using a range of artificial intelligence-based software, with the intention of establishing a multi-dimensional design space that ensures consistent product quality. At the end of the study, a design space was developed based on the study data and specifications, and a new formulation was optimized. On the basis of this formulation, a new laboratory batch formulation was prepared and tested. It was confirmed that the explored formulation was within the design space.     

失效模式和影响分析在原料药工艺验证中的应用

目的探索风险管理工具在原料药工艺验证中的应用,使用失效模式和影响分析(failure mode and effects analysis,FMEA)评估关键工艺参数,降低验证风险。方法将FMEA应用于原料药的工艺验证,根据风险优先数(risk priority number,RPN)值的大小,确定关键工艺参数和风险控制措施。结果通过实施和跟踪工艺操作控制措施,再次计算RPN值,较验证前减小,降低发生工艺偏差的风险。结论质量风险管理应用于原料药工艺验证,可有效地提高针对性,降低生产质量的系统风险,将有利于日常商业化生产的平稳运行,提高生产效率。     

马尾松树皮提取物体外抑制人肝癌细胞BEL-7402生长的数学建模

采用三因素五水平二次通用旋转组合设计 ,研究马尾松树皮提取物 (Pinus massaniana bark extract,PMBE)、胎牛血清 (fetal bovine serum,FBS)和时间 (t)对体外培养人肝癌细胞 BEL-740 2生长的影响 ,建立了三因素对体外培养 BEL -740 2细胞生长抑制率的反应模型 ,并利用回归模型对三因素优化组合 ,同时就各因素单独效应及其互作效应进行了探讨。三因素适量搭配可提高BEL-740 2的生长抑制率 ,最高可达 0 .3 6,此时三者用量水平的最佳组合为 (PMBE,FBS,t) =(0 ,0 ,-1.682 ) ,即三因子用量分别为 160μg/ml、10 %、2 7.816h时抑制率最高     

Contribution of Monoamine Oxidase (MAO) to the Binding of Tertiary Basic Drugs in Lung Mitochondria

The effects of tertiary amine-containing basic drugs on the enzymes located in the mitochondria and the effect of monoamine oxidase inhibitors (MAOIs) on drug accumulation in lung mitochondria have been studied. Various basic drugs inhibited MAO activity but not other mitochondrial marker enzymes. The potency of MAO inhibition correlated well with their lipid solubility, and the basic drugs inhibited MAO activity dose dependency and competitively. Further, MAO inhibition correlated well with binding affinity to lung mitochondria, and the binding of tertiary amine drugs to lung mitochondria was decreased by treatment with MAOIs. A good correlation was observed between the potency of MAOIs to inhibit the binding of the basic drug to the high-affinity site in mitochondria and the MAO inhibitory activity in mitochondria. These results indicate that mitochondrial MAO is one of the binding sites for tertiary basic drugs in the lung. We think that the action and/or adverse reaction of some drugs may result from inhibition of mitochondrial MAO to metabolize various biogenic amines and that mitochondrial MAO may function as a reservoir for basic drugs.     

Application of the threshold of toxicological concern (TTC) to the safety evaluation of cosmetic ingredients

The threshold of toxicological concern (TTC) has been used for the safety assessment of packaging migrants and flavouring agents that occur in food. The approach compares the estimated oral intake with a TTC value derived from chronic oral toxicity data for structurally-related compounds. Application of the TTC approach to cosmetic ingredients and impurities requires consideration of whether route-dependent differences in first-pass metabolism could affect the applicability of TTC values derived from oral data to the topical route. The physicochemical characteristics of the chemical and the pattern of cosmetic use would affect the long-term average internal dose that is compared with the relevant TTC value. Analysis has shown that the oral TTC values are valid for topical exposures and that the relationship between the external topical dose and the internal dose can be taken into account by conservative default adjustment factors. The TTC approach relates to systemic effects, and use of the proposed procedure would not provide an assessment of any local effects at the site of application. Overall the TTC approach provides a useful additional tool for the safety evaluation of cosmetic ingredients and impurities of known chemical structure in the absence of chemical-specific toxicology data.     

Application of a Biomagnetic Measurement System (BMS) to the Evaluation of Gastrointestinal Transit of Intestinal Pressure-Controlled Colon Delivery Capsules (PCDCs) in Human Subjects

Purpose. For determination of the transit time through various partsof the gastrointestinal (GI) tract, we developed a method that providesthe location of disintegration and drug release. This method involves GImagnetomarkergraphy (GIMG) using a 129-channel Shimadzu vectorbiomagnetic measurement system (BMS). Methods. To magnetically label the pressure-controlled colon deliverycapsule (PCDC) containing 75.0 ± 0.5 mg of caffeine as a tracer drug,small capsule caps containing 90 mg of ferric oxide powdered magnetite(Fe₂O₃) were attached to PCDCs. After orally administration to fastedhuman volunteers, saliva samples were collected hourly and salivarycaffeine concentration was measured. At the same time, locations ofthe magnetic PCDC were detected by BMS just after the PCDCs weremagnetized with the coils of a magnetic resonance imaging (MRI)system. The magnetic field distributions were analyzed and theestimated positions were shown on the MRI picture of the same subject'sabdominal structure. Results. We magnetized PCDC with permanent magnets or anelectromagnet before ingestion and the estimated locations of PCDC in the GItract exhibited high estimation error. In order to increase the precision ofestimated localization of PCDCs, PCDCs were magnetized within thecoils of the MRI. As a result, these PCDCs had strong magnetic dipolesthat were parallel to the sensor unit of BMS in every measurement,and therefore the spatial resolution of the PCDC's two-dimensionalpositions in the organs of the GI tract was within a range of severalmillimeters. Conclusions. GIMG is a powerful tool for the study of colon deliveryefficiencies of PCDCs. The main advantage of GIMG is the capabilityto obtain even more detailed knowledge of the behavior and fate ofsolid pharmaceutical formulations during GI passage.     

神经网络用于苯乙胺衍生物的QSAR研究

将神经网络应用于定量构效关系（ＱＳＡＲ）中。采用改进的反向传播算法探讨了肾上腺素能阻断剂Ｎ，Ｎ－二甲基－２－溴苯乙胺取代衍生物的生物活性与取代基疏水参数（Σπ）和电子参数（Σσ）之间的关系。获得了精密的拟合及准确的预测（最大误差小于１０％），优于多无线性回归法。作为一种有效的化学计量学工具（Ｃｈｅｍｏｍｅｔｒｉｃｓ），神经网络具有良好的预测效果及较强的非线性处理功能，可望在ＱＳＡＲ及药物制剂中发挥重要作用。