Application of arti ficial intelligence in modeling of the doxorubicin release behavior of pH and temperature responsive poly(NIPAAm-co-AAc)-PEG IPN hydrogel


BOZTEPE C. , KÜNKÜL A. , Yuceer M.

JOURNAL OF DRUG DELIVERY SCIENCE AND TECHNOLOGY, cilt.57, 2020 (SCI İndekslerine Giren Dergi) identifier identifier

  • Cilt numarası: 57
  • Basım Tarihi: 2020
  • Doi Numarası: 10.1016/j.jddst.2020.101603
  • Dergi Adı: JOURNAL OF DRUG DELIVERY SCIENCE AND TECHNOLOGY

Özet

Modeling of the drug release behavior of stimuli-responsive hydrogels is a domain of steadily increasing academic and industrial importance. It is very difficult to accurately predict the drug release kinetic of this type drug carrier materials due to environmental variables in the body such as pH and temperature. In this study, a pH- and temperature-responsive poly(N-Isopropyl acrylamide-co-Acrylic acid)/Poly(ethylene glycol) (poly(NIPAAm-co-AAc)/PEG) interpenetrating polymer network (IPN) hydrogel was synthesized by free radical solution polymerization in the presence of poly(NIPAAm-co-AAc) microgels and PEG. The synthesized IPN hydrogels showed rapid pH- and temperature-responsive deswelling behavior. The textural properties and surface morphology of poly(NIPAAm-co-AAc) IPN hydrogel were characterized by scanning electron microscopy (SEM) analysis technique. The doxorubicin (DOX) was loaded to the hydrogels by swelling the hydrogels in the DOX solution. The cumulative release of DOX has been investigated as a function of time, pH and temperature. Experimental DOX release data obtained were successfully modeled using ANNs, LS-SVM and SVR methodologies. To evaluate the performance of these models, four statistical parameters: correlation coefficient (R), root mean squared error (RMSE), mean square error (MSE) and mean absolute percentage error (MAPE) were calculated. It was found that the developed ANN model show best performance in modeling the DOX release behavior of poly(NIPAAm-co-AAc)/PEG IPN hydrogels.