Neural Networks Used for Modeling of Processes Involving Pharmaceutical Polymers

1 Neural Networks Used for Modeling of Processes Involving Pharmaceutical Polymers 

The aim of this approach [11] was to develop a simple, nevnal network model of ketoprofen dissolution from solid dispersions (SD) and physical mixtures (PM). This model could be an aid in prospective formulation of solid dispersions with ketoprofen, predicting its dissolution characteristics from unknown pharmaceutical formulations. The seven inputs were chosen for the model formulation of SD or PM, SD previously mixed or not, method of preparation (evaporation or melting), cooling on ice or ambient temperature, amount of ketoprofen in SD or PM formulation, carrier connectivity index and time of dissolution test the output was the dissolved ketoprofen at a specified time in the input. 

Regarding the distribution of generalization errors, it has been found that around 75% of the predictions were in acceptable range, which can be considered a reasonable system performance. 

Throughout this article, other examples of ANNs used in the formulation of various dosage forms can be enumerated microemulsions [12], lipophilic semisolid emulsion [13], and hydrogels [14]. 

There are many other examples of applications of ANNs in pharmaceutical technology, dted in Sun et al. [ 16]. It has been shown that many artificial intelligence systems, especially neural networks, can be applied to the fundamental investigations of the effects of formulation and process variables on the delivery system. 

---