The Use of Data Mining for Obtaining Deeper Insights into the Fabrication of Prednisolone-Loaded Chitosan Nanoparticles

Jehad Nasereddin,Reem Al Wadi,Ahlam Zaid Al-Kilani,Asad Abu Khalil,Mohammad Al Natour,Wael Abu Dayyih
DOI: https://doi.org/10.1208/s12249-024-02756-3
IF: 3.3
2024-02-16
AAPS PharmSciTech
Abstract:The present work explores a data mining approach to study the fabrication of prednisolone-loaded chitosan nanoparticles and their properties. Eight PLC formulations were prepared using an automated adaptation of the antisolvent precipitation method. The PLCs were characterized using dynamic light scattering, infrared spectroscopy, and drug release studies. Results showed that that the effective diameter, loading capacity, encapsulation efficiency, zeta potential, and polydispersity of the PLCs were influenced by the concentration and molecular weight of chitosan. The drug release studies showed that PLCs exhibited significant dissolution enhancement compared to pure prednisolone crystals. Principal components analysis and partial least squares regression were applied to the infrared spectra and the DLS data to extract higher-order interactions and correlations between the critical quality attributes and the diameter of the PLCs. Principal components revealed that the spectra clustered according to the type of material, with PLCs forming a separate cluster from the raw materials and the physical mix. PLS was successful in predicting the ED of the PLCs from the FTIR spectra with R 2 = 0.98 and RMSE = 27.18. The present work demonstrates that data mining techniques can be useful tools for obtaining deeper insights into the fabrication and properties of PLCs, and for optimizing their quality and performance. It also suggests that FTIR spectroscopy can be a rapid and non-destructive method for predicting the ED of PLCs.
pharmacology & pharmacy
What problem does this paper attempt to address?