Modelling of the solvent evaporation method for the preparation of controlled release acrylic microspheres using neural networks


Yuksel N., Turkoglu M., Baykara T.

JOURNAL OF MICROENCAPSULATION, cilt.17, sa.5, ss.541-551, 2000 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 17 Sayı: 5
  • Basım Tarihi: 2000
  • Dergi Adı: JOURNAL OF MICROENCAPSULATION
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.541-551
  • Anahtar Kelimeler: solvent evaporation, acrylic microspheres, preparative variables, artificial neural networks, multiple regression, PHARMACEUTICAL PRODUCT DEVELOPMENT, BED GRANULATION PROCESS, POLYMERIC MICROSPHERES, MICROCAPSULES, INDOMETHACIN
  • Marmara Üniversitesi Adresli: Hayır

Özet

The purpose of the present study was to model the solvent evaporation procedure for the preparation of acrylic microspheres by using artificial neural networks (ANNs) to obtain an understanding of the selected preparative variables. Three preparative variables, the concentration of the dispersing agent (sucrose stearate), the stirring rate of emulsion system, and the ratio of polymers (Eudragit RS-L) were studied, each at different levels, as input variables. The response (output) variables examined to characterize microspheres and drug release were the size of the microspheres and T-63.2%, the time at which 63.2% of drug is released. The results were also analysed by the multiple linear regression (MLR) to provide a comparison with the ANN methodology. Although both ANN and MLR methods were found to be similar in characterizing the process studied, the results showed that an ANN method gave a better prediction than the MLR method. For the size values of the microspheres, the predictability of the ANN model was quite high (R-2 = 0.9602) based on the input variables. A relationship between these variables and size values of microspheres was also obtained by the MLR model (R-2 = 0.9602). The performances of both models for the release data from microspheres based on the same input variables were at the level of 53%. According to the results, the ANN methodology can provide an alternative to the traditional regression methods, as a flexible and accurate method to study process and formulation factors.