Early pharmacokinetic optimisation is a key principle in drug discovery and development. Modeling absorption, distribution, metabolism and excretion (ADME) using experimentally-derived data is time-consuming and expensive. The use of computational in silico techniques to predict pharmacokinetic properties based on molecular structure is gaining wider validity and acceptance in the pharmaceutical industry. This book describes the use of artificial neural networks (ANN) as robust nonlinear modeling tools for developing quantitative structure-pharmacokinetic relationships (QSPkR). Different ANN paradigms are examined for predictive modeling of various pharmacokinetic parameters, both individually and simultaneously. Consideration is given to physiological processes, drug and molecular structural data, and model interpretation. As well as providing the theory behind ANN model construction, this book details their practical application in pharmaceutical research and gives meaning to many of the theoretically-derived molecular descriptors now available. A valuable resource for medicinal chemists and pharmaceutical scientists engaging in structure-property and structure-activity modeling.