modeling the parameters involved in preparation of pla nanoparticles carrying hydrophobic drug molecules using artificial neural networks

artificial neural networks (anns) are used to optimize a formulation of poly(lactic acid) (pla) nanoparticles containing hydrophobic drug molecules through a study of the critical parameters affecting nanoparticle size.

methods

we evaluate the effect of input variables, including concentrations of pla and tween 80, amplitude of ultrasound wave, and sonication time on the formation of pla nanoparticles, which were prepared using a solvent evaporation method. budesonide was used as a model hydrophobic drug. an ann model was created using training data and evaluated for prediction capability using validation data.

results

the ann model demonstrated that reducing pla concentration and increasing tween 80 concentration provided optimum conditions for the preparation of small particle size. additionally, the simultaneous use of high sonication time and amplitude has an adverse effect on particle diameter.

conclusion

by defining the effects of each parameter on the size of pla nanoparticles, this study demonstrated the feasibility of using an ann model to optimize the conditions for achieving minimum particle size in hydrophobic drug-loaded pla nanoparticles.