Artificial neural network for optimizing the formulation of curcumin-loaded liposomes from statistically designed...

Area of Science:

• Pharmacology and Pharmaceutical Sciences
• Biomaterials Science
• Computational Chemistry

Background:

• Curcumin, a polyphenol from Curcuma Longa, exhibits beneficial effects on skin regeneration and wound healing by modulating inflammatory and proliferative cellular processes.
• Curcumin's lipophilic nature necessitates advanced drug delivery systems for effective therapeutic application.
• Liposomes, biocompatible lipid vesicles, are suitable for encapsulating curcumin but require careful formulation to ensure stability and high encapsulation efficiency.

Purpose of the Study:

• To optimize the formulation of curcumin-loaded liposomes to enhance encapsulation efficiency and drug flux.
• To investigate the application of artificial neural networks (ANN) for predicting and optimizing liposome formulation parameters.
• To establish statistically designed models for reproducible and effective curcumin liposome preparation.

Main Methods:

• Artificial neural network (ANN) modeling was employed to optimize curcumin-loaded liposome formulations.
• Key formulation parameters including lipid/curcumin ratio, sonication time, and hydration volume were systematically varied.
• Encapsulation efficiency and drug flux were measured as response variables to assess formulation performance.

Main Results:

• The optimized formulation, determined by ANN, utilized a lipid/curcumin ratio of 4.35, 15 minutes of sonication, and 25 mL hydration volume.
• Predicted maximum encapsulation efficiency reached 100%, with a flux of 51.23 µg/cm² /h.
• Experimental validation closely matched predicted values, achieving 99.934% encapsulation efficiency and 51.229 µg/cm²/h flux.

Conclusions:

• Artificial neural network-driven optimization significantly enhances curcumin-loaded liposome formulation for improved drug delivery.
• Optimized curcumin-loaded liposomes demonstrate high encapsulation efficiency and flux, indicating their potential as a promising pharmaceutical product.
• Statistically designed models derived from ANN are crucial for developing stable and effective liposomal drug delivery systems in the pharmaceutical industry.