Basic concepts of artificial neural network (ANN) modeling and its application in pharmaceutical research

Artificial neural networks (ANNs) are biologically inspired computer progra ms designed to simulate the way in which the human brain processes informat ion. ANNs gather their knowledge by detecting the patterns and relationship s in data and learn (or are trained) through experience, not from programmi ng. An ANN is formed from hundreds of single units, artificial neurons or p rocessing elements (PE)I connected with coefficients (weights), which const itute the neural structure and are organised in layers. The power of neural computations comes from connecting neurons in a network. Each PE has weigh ted inputs, transfer function and one output. The behavior of a neural netw ork is determined by the transfer functions of its neurons, by the learning rule, and by the architecture itself. The weights are the adjustable param eters and, in that sense, a neural network is a parameterized system. The w eighed sum of the inputs constitutes the activation of the neuron. The acti vation signal is passed through transfer function to produce a single outpu t of the neuron. Transfer function introduces non-linearity to the network. During training, the inter-unit connections are optimized until the error in predictions is minimized and the network reaches the specified level of accuracy. Once the network is trained and tested it can be given new input information to predict the output. Many types of neural networks have been designed already and new ones are invented every week but all can be descri bed by the transfer functions of their neurons, by the learning rule, and b y the connection formula. ANN represents a promising modeling technique, es pecially for data sets having non-linear relationships which are frequently encountered in pharmaceutical processes. In terms of model specification, artificial neural networks require no knowledge of the data source but, sin ce they often contain many weights that must be estimated, they require lar ge training sets. In addition, ANNs can combine and incorporate both litera ture-based and experimental data to solve problems. The various application s of ANNs can be summarised into classification or pattern recognition, pre diction and modeling. Supervised associating networks can be applied in pha rmaceutical fields as an alternative to conventional response surface metho dology. Unsupervised feature-extracting networks represent an alternative t o principal component analysis. Non-adaptive unsupervised networks are able to reconstruct their patterns when presented with noisy samples and can be used for image recognition. The potential applications of ANN methodology in the pharmaceutical sciences range from interpretation of analytical data , drug and dosage form design through biopharmacy to clinical pharmacy. (C) 2000 Elsevier Science B.V. All rights reserved.