Formulation optimization technique based on artificial neural network in salbutamol sulfate osmotic pump tablets.

Area of Science:

• Pharmaceutical Sciences
• Computational Chemistry
• Drug Delivery Systems

Background:

• Optimizing pharmaceutical formulations is crucial for predictable drug release.
• Artificial neural networks (ANNs) offer potential for complex system modeling.

Purpose of the Study:

• To develop and validate an artificial neural network (ANN) based technique for optimizing salbutamol sulfate osmotic pump tablet formulations.
• To assess the ANN's ability to predict drug release profiles based on formulation variables.

Main Methods:

• Preparation of 30 salbutamol sulfate osmotic pump tablet formulations with varying hydroxypropyl methylcellulose (HPMC), polyethylene glycol 1500 (PEG1500), and coat weight.
• Dissolution testing to obtain drug release parameters (average release rate 'v' and correlation coefficient 'r').
• Training and testing an ANN model using formulation factors and release parameters.

Main Results:

• The ANN model was successfully trained, achieving a satisfactory error function value for test data.
• The optimal formulation predicted by the ANN exhibited a release profile that closely matched observed results.
• The developed technique demonstrated accurate prediction of drug release characteristics.

Conclusions:

• Artificial neural networks (ANNs) are effective tools for optimizing pharmaceutical formulations.
• This ANN-based approach can reliably predict drug release profiles, aiding in efficient formulation development.
• The study validates the use of ANNs in achieving desired drug release characteristics for osmotic pump systems.