Artificial Intelligence:: Application of Neural Networks, Genetic Algorithms and Fuzzy Logic in the Formulation of Ramipril, Hydrochlorothiazide, Caffeine, Diclofenac Sodium and Ondansetron
Abstract:
Pharmaceutical drug manufacturing is a complicated process. The interactions between raw materials and process conditions are very important for the quality of the finished product. The use of artificial intelligence provides better understanding of relationships between different formulation and process parameters.
Computerized systems such as Artificial Neural Networks, Genetic Algorithms and Fuzzy Logic are required to design experimental space for the required data. Artificial Neural Network (ANN) is based on an experimental model of the data-processing methods of a biological brain. ANNs are capable of understanding cause and effect relationships between inputs and outputs. The Genetic Algorithm is useful in predicting the results that arise from changes in input parameters.
Fuzzy Logic generates interpretable rules from complex and non-linear data. The modeling of ramipril tablet formulation, Direct compression tablet formulation containing Hydrochlorothiazide, Formulation of Caffeine, Diclofenac sodium SR matrix tablets and Ondansetron orally disintegrating tablets using the new science and risk-based techniques has many advantages over traditional modeling techniques.
