Application of lignin in controlled release: development of predictive model based on artificial neural network for API release
Predictive models for simulation of drug release from tablets containing lignin as excipient were developed in this work. Two predictive models including Artificial Neural Network (ANN) and hybrid ANN-Kriging were developed to simulate the tablet dissolution. Measured data was collected on the release rate of aspirin tablets prepared by dry granulation via roll compaction followed by milling and tableting. Two formulations were considered, one with lignin and one without. The main aim is to show the effect of lignin as a bio-based natural polymer in tablet manufacturing to control drug dissolution. For the ANN model development, process and formulation parameters including roll pressure and lignin content were considered as the input, while API dissolution was considered as response. The predictions were compared with measured data to calibrate and validate the model. To improve the predictability of the model, Kriging interpolation was used to enhance the number of training points for the ANN. The interpolated data was trained and validated. The final concentration and the dissolution rate were predicted by ANN as well as ANN-Kriging models, and the R2 of greater than 0.99 for most cases was obtained. The validated model was used to evaluate the effect of process parameters on the release rate and it was indicated that the tablets containing lignin have higher release rate compared to tablets without. Also, it was revealed that process parameters do not have significant effect on the tablet release rate, and the tablet release rate is mainly affected by the lignin content. The results indicated that ANN-based model is a powerful tool to predict the API release rate for tablets containing various formulations, and can be used as a predictive tool for design of controlled release systems.
KeywordsANN model Kriging interpolation Controlled release Lignin Pharmaceuticals Tableting Dry granulation
This research was conducted with the financial support of the Synthesis and Solid State Pharmaceutical Centre (SSPC), funded by SFI and is co-funded under the European Regional Development Fund under Grant Number 14/SP/2750.
Conflict of interest
The authors declare that they have no conflict of interest.
- Barrasso D, El Hagrasy A, Litster JD, Ramachandran R (2015) Multi-dimensional population balance model development and validation for a twin screw granulation process. Powder Technol 270, Part B, 612–621Google Scholar
- Ismail HY, Singh M, Darwish S, Kuhs M, Shirazian S, Croker DM, Khraisheh M, Albadarin AB, Walker GM (2019b) Developing ANN-Kriging hybrid model based on process parameters for prediction of mean residence time distribution in twin-screw wet granulation. Powder Technol 343:568–577CrossRefGoogle Scholar
- SAS_Institute (2016) Using JMP® 12.2.0, Cary, NCGoogle Scholar