Oral disintegrating tab lets (ODTs) are a nov el dosage form that can be dissolved on the tonguewithin 3 min or less especially for geriatric and pediatric patients. Current ODT formula-tion studies usually rely on the personal experience of pharmaceutical experts and trial-and-error in the laboratory, which is inefficient and time-consuming. The aim of currentresearch was to establish the prediction model of ODT formulations with direct compres-sion pr ocess by artificial neural network (ANN) and dee p neural network (DNN) techniques.145 formulation data were extracted from Web of Science. All datasets were divided intothree parts: training set (105 data), validation set (20) and testing set (20). ANN and DNNwere compared for the prediction of the disinte grating time.The accuracy of the ANN modelhave reached 85.60%, 80.00% and 75.00% on the training set, validation set and testing setrespectively, whereas that of the DNN model were 85.60%, 85.00% and 80.00%, respectively.Compared with the ANN, DNN showed the better prediction for ODT formulations. It is thefirst time that deep neural network with the improv ed dataset selection algorithm is appliedto formulation prediction on small data. The proposed predictive approach could evaluatethe critical parameters about quality control of formulation, and guide research and processdevelopment. The implementation of this prediction model could effectively reduce drugproduct development timeline and material usage, and proactively facilitate the develop-ment of a robust drug product

Aim

The aim of current research was to establish the prediction model of ODT formulations with direct compression process by artificial neural network (ANN) and deep neural network (DNN) techniques.

Method

145 formulation data were extracted from Web of Science. All data sets were divided into three parts: training set (105), validation set (20) and testing set (20). Artificial neural network (ANN) and deep neural network (DNN) were compared for the prediction of the disintegrating time.

Results

The accuracy of the ANN modelhave reached 85.60%, 80.00% and 75.00% on the training set, validation set and testing setrespectively, whereas that of the DNN model were 85.60%, 85.00% and 80.00%, respectively.Compared with the ANN, DNN showed the better prediction for ODT formulations.

Conclusion

It is the first time that DNN with the improved dataset selection algorithm is applied to formulation prediction on small data. The proposed predictive approach could evaluate the critical parameters about quality control of formulation, and guide research and process development. The implementation of this prediction model could effectively reduce drug product development timeline and material usage, and proactively facilitate the development of a robust drug product.

Publication

Run Han, Yilong Yang, Xiaoshan Li, Defang Ouyang, Predicting oral disintegrating tablet formulations by neural network techniques, Asian Journal of Pharmaceutical Sciences, 2018.

Source code

The source code is available at GitHub Click here