Applied Artificial Neural Networks: from Associative Memories to Biomedical Applications

Due to remarkable capabilities of artificial neural networks (ANNs) such as generalization and nonlinear system modeling, ANNs have been extensively studied and applied in a wide variety of applications (Amiri et al., 2007; Davande et al., 2008). The rapid development of ANN technology in recent years has led to an entirely new approach for the solution of many data processing-based problems, usually encountered in real applications (Hosseini et al., 2007). ANNs are characterized in principle by a network topology, a connection pattern, neural activation properties and a training strategy to process data. In this section, a brief explanation for four types of ANNs including Multi-Layer Perceptron (MLP), Radial Basis Function Network (RBFN), Generalized Regression Neural Network (GRNN) and SelfFeedback Neural Network (SFNN) is provided. The MLP, RBFN and GRNN belong to a feed-forward class of neural networks (FFNN), while the SFNN belongs to the other important class of neural networks that is recurrent neural networks (RNN). Next, we investigate the diverse and innovative applications of these neural networks such as associative neural networks for recognition of analog and digital patterns, estimating the release profile of betamethasone (BTM) and betamethasone acetate (BTMA) and optimization of drug delivery system formulation. Regarding the first application, we propose a hybrid model consists of the SFNN in parallel with the GRNN. In the proposed hybrid model, storing of desired patterns is performed by employing a new one-shot learning algorithm put forward in the chapter. It will be shown that this new hybrid model is able to perform essential properties found in associative memories such as generalization, completion and recognition of corrupted patterns. Moreover, a number of case studies are performed for the purpose of performance comparison between the hybrid model and others from different classes. For the recurrent associative memory class, the comparison is