Formula optimization of theophylline controlled-release tablet based on artificial neural networks

doi:10.1016/S0168-3659(00)00248-0

Journal of Controlled Release

Volume 68, Issue 2, 10 August 2000, Pages 175-186

https://doi.org/10.1016/S0168-3659(00)00248-0 Get rights and content

Abstract

Formulation of the controlled-release tablet containing theophylline was optimized based on the simultaneous optimization technique in which an artificial neural network (ANN) was incorporated. As model formulations, 16 kinds of theophylline tablets were prepared. The amounts of Controse, the mixture of hydroxypropylmethyl cellulose with lactose, cornstarch and compression pressure were selected as causal factors. The release profiles of theophylline were characterized as the sum of the fast and slow release fractions. The initial weight, the rate constant in the fast release fraction and the rate constant in the slow release fraction were estimated as release parameters. A set of release parameters and causal factors were used as tutorial data for ANN and analyzed using a computer. Based on the plasma concentration profiles of theophylline predicted by the pharmacokinetic analysis in humans, a desirable set of release parameters was provided. The simultaneous optimization was performed by minimizing the generalized distance between the predicted values of each response and the desirable one that was optimized individually. The optimization technique incorporating ANN showed a fairly good agreement between the observed values of release parameters and the predicted results.

Introduction

In the design of pharmaceutical formulations, it is important to optimize pharmaceutical responses relating to its effectiveness, safety and usefulness to patients. A computer optimization technique, based on a response surface method (RSM), has often been used for selecting pharmaceutical formulations [1], [2]. The optimization procedure based on RSM includes statistical experimental designs, multiple regression analysis and mathematical optimization algorithms for seeking the best formulation under a set of constrained equations. Since theoretical relationships between response variables and causal factors are not clear, multiple regression analysis can be applied to the prediction of response variables on the basis of a second-order polynomial equation. The prediction of pharmaceutical responses based on the second-order polynomial equation, however, is often limited to low levels, resulting in the poor estimation of optimal formulations. To overcome the shortcomings in RSM, we developed a multi-objective simultaneous optimization technique in which an artificial neural network (ANN) was incorporated [3], [4], [5]. ANN is a learning system based on a computational technique which can simulate the neurological processing ability of the human brain [6], and could be applied to quantifying a non-linear relationship between causal factors and pharmaceutical responses by means of iterative training of data obtained from a designed experiment.

The aim of the present study was to apply the simultaneous optimization method incorporating ANN to the development of a theophylline tablet, which has an optimized release behavior. Plasma concentrations of theophylline were predicted on the basis of pharmacokinetic analysis to obtain desirable release profiles of theophylline from the tablets. Controse, the mixture of hydroxypropylmethylcellulose (HPMC) and lactose, was employed as a gel-forming material while cornstarch was used as a disintegrant.

Section snippets

ANN architecture

Theoretical details of the hierarchical ANN have been given elsewhere. Briefly, the general structure of ANN has one input layer, one or more hidden layers and one output layer. Each layer has some units corresponding to neurons. The units in neighboring layers are fully interconnected with links corresponding to synapses. The strengths of connections between two units are called ‘weights’. ANN learns an approximate non-linear relationship by a training procedure, which involves varying weight

Materials and methods

Theophylline was purchased from Sigma–Aldrich (Tokyo, Japan). Controse 80/20 was generously supplied by Freund (Tokyo, Japan). Controse 80/20 is a powder mixture of HPMC and lactose in the weight ratio of 80 to 20, passing through mesh No. 45 in JIS gauge. Cornstarch, JP grade, was purchased from Kosakai (Tokyo, Japan).

Release behaviors of theophylline

Release behaviors of theophylline from the tablets in Table 1 are shown in Fig. 2. The fast release rate of theophylline during the initial stage and consecutive slow release in the following stage were observed. The tablets containing large amounts of cornstarch were partly disintegrated during the initial stage and then a hydrogel was gradually formed on the surface of the tablets. However, the initial disintegration was hardly observed in the tablets containing small amounts of cornstarch.

Acknowledgements

This work was supported by the Uehara Memorial Foundation.

References (18)

J. Takahara et al.
Multi-objective simultaneous optimization based on artificial neural network in sustained release formulations
J. Control. Rel.
(1997)
J. Takahara et al.
Multi-objective simultaneous optimization based on artificial neural network in a ketoprofen hydrogel formula containing O-ethylmenthol as a percutaneous absorption enhancer
Int. J. Pharm.
(1997)
K. Takayama et al.
Simultaneous optimization for several characteristics concerning percutaneous absorption and skin damage of ketoprofen hydrogels containing d-limonene
Int. J. Pharm.
(1991)
A.I. Khuri et al.
Response Surface, Design and Analysis
(1987)
G.A. Lewis et al.
Pharmaceutical Experimental Design
(1999)
K. Takayama et al.
Artificial neural network as a novel method to optimize pharmaceutical formulations
Pharm. Res.
(1999)
A.S. Achanta et al.
Artificial neural networks: Implications for pharmaceutical sciences
Drug Dev. Ind. Pharm.
(1995)
J.L. McClelland et al.
Explorations in Parallel Distributed Processing
(1988)
R.J. Erb
Introduction to backpropagation neural network computation
Pharm. Res.
(1993)

There are more references available in the full text version of this article.

Cited by (90)

Artificial intelligence applications in pharmaceutical technology
2023, Computer-Aided Applications in Pharmaceutical Technology: Delivery Systems, Dosage Forms, and Pharmaceutical Unit Operations
This chapter presents a review of the possible applications of methods based on neural computing in pharmaceutical products and process development. Some of the methods described are used for classification purposes, whereas others can be applied to modeling and optimization, or even induction of rules. The basic concepts of each method are theoretically described, followed by examples of their application in pharmaceutical technology. A theoretical background aims to provide a better understanding of the methods and is based on their most important features. Examples should encourage the reader to embrace the abovementioned methods and use them to complement conventional statistical methods for classification and regression.
AI approaches for the development of drug delivery systems
2023, A Handbook of Artificial Intelligence in Drug Delivery
The pharmaceutical industry is facing challenges in sustaining its drug development programs because of increased R&D costs and reduced efficiency. Remarkable improvements in computational power coupled with advancements in artificial intelligence (AI) technology could be utilized to revolutionize the drug development process. Machine learning (ML) is a subfield of AI, which uses statistical methods with the ability to learn with or without being explicitly programmed. A further subfield of ML called deep learning (DL) uses artificial neural networks (ANNs), which is a multiobjective simultaneous optimization technique of drug delivery systems. ANNs are used in pharmaceutical and pharmacokinetic areas to model complex relationships and to predict the nonlinear relationship between causal factors and response variables. The distinct features of an ANN make it extremely useful in situations in which the functional correlation between inputs and outputs is unclear. This chapter covers AI in general and its applications in the drug delivery field with special emphasis on novel drug delivery systems.
AI in microfabrication technology
2023, A Handbook of Artificial Intelligence in Drug Delivery
There has been recent interest in the application of artificial intelligence (AI) in drug delivery systems. One of the areas of development includes implantable, smart devices designed for targeted delivery which can be programmed to release drugs in a specific quantity at a specific time. This delivery can be automatic or triggered from external stimuli, effectively being delivered by hybrid intelligent systems. The fabrication of these devices has been a crucial recent advance, allowing for improvement of a variety of methods of intelligent drug delivery, including microelectromechanical systems (MEMS), implantable drug reservoirs, 3D printing, nanobots, microfluidic platforms, organ-on-a-chip (OOC), and smart biosensors.
Applications of Machine Learning in Solid Oral Dosage Form Development
2021, Journal of Pharmaceutical Sciences
This review comprehensively summarizes the application of machine learning in solid oral dosage form development over the past three decades. In both academia and industry, machine learning is increasingly applied for multiple preformulation/formulation and process development studies. Further, this review provides the authors’ perspectives on how pharmaceutical scientists can use machine learning for right projects and in right ways; some key ingredients include (1) the determination of inputs, outputs, and objectives; (2) the generation of a database containing high-quality data; (3) the development of machine learning models based on dataset training and model optimization; (4) the application of trained models in making predictions for new samples. It is expected by the authors and others that machine learning will promisingly play a more important role in tomorrow's projects for solid oral dosage form development.
Potential of Raman spectroscopy in facilitating pharmaceutical formulations development – An AI perspective
2021, International Journal of Pharmaceutics
Drug development is time-consuming and inherently possesses a high failure rate. Pharmaceutical formulation development is the bridge that links a new chemical entity (NCE) to pre-clinical and clinical trials, and has a high impact on the efficacy and safety of the final drug product. Further, the time required for this process is escalating as formulation techniques are becoming more complicated due to the rising demands for drug products with better efficacy and patient compliance, as well as the inherent difficulties of addressing the unfavorable properties of NCEs such as low water solubility. The advent of artificial intelligence (AI) provides possibilities to accelerate the drug development process. In this review, we first examine applications of AI methods in different types of pharmaceutical formulations and formulation techniques. Moreover, as availability of data is the engine for the advancement of AI, we then suggest a potential way (i.e. applying Raman spectroscopy) for faster high-quality data gathering from formulations. Raman techniques have the capability of analyzing the composition and distribution of components and the physicochemical properties thereof within formulations, which are prominent factors governing drug dissolution profiles and subsequently bioavailability. Thus, useful information can be obtained bridging formulation development to the final product quality.
Applications of Computers in Pharmaceutical Product Formulation
2018, Dosage Form Design Parameters
The traditional way of discovery, development, and formulation of a new drug is tedious, highly time-consuming, as well as costly with high attrition rates. However, with continuous advancements of in silico technologies and algorithms, the virtual screening strategies have emerged as reliable and unavoidable alternative or complementary approaches to conventional methods at the different stages of the discovery and development of drugs until they reach the market. Current sophistication of in silico approaches not only aid in the speed of the process but also promoting the design, discovery, and optimization of new chemical entities (NCEs) more rationally while reducing the attrition rate at the later stages of development and formulation of drugs. This chapter mainly focuses and discusses the basic concepts, methodology, applications, and pitfalls of various computer-aided expert systems, artificial neural networking (ANN), and computer simulation strategies that are presently being used for the design and development of pharmaceutical formulations. An attempt has also been made to contextualize the chapter with an overview of the role of various in silico structure-based drug design approaches (molecular docking and molecular dynamics simulation) and ligand-based drug design strategies (quantitative structure–activity relationship (QSAR) and pharmacophore modeling) that are being currently used in order to gain insight into the binding interactions of biological target proteins with their substrates or drug molecules for the design, discovery, and optimization of lead molecules along with the prediction of pharmacokinetic (absorption, distribution, metabolism, excretion, and toxicity—ADMET) properties.

View all citing articles on Scopus

View full text

Formula optimization of theophylline controlled-release tablet based on artificial neural networks

Abstract

Introduction

Section snippets

ANN architecture

Materials and methods

Release behaviors of theophylline

Acknowledgements

J. Control. Rel.

Int. J. Pharm.

Int. J. Pharm.

Response Surface, Design and Analysis

Pharmaceutical Experimental Design

Artificial neural network as a novel method to optimize pharmaceutical formulations

Pharm. Res.

Artificial neural networks: Implications for pharmaceutical sciences

Drug Dev. Ind. Pharm.

Explorations in Parallel Distributed Processing

Introduction to backpropagation neural network computation

Pharm. Res.