Optimization of dropping process of Xuesaitong dropping pills based on quality by design concept and machine vision

Yizhe Hou,Xi Wang,Zhiyong Zhang,Jiaheng Wu,Xiang Cai,Pian Li,Zheng Li,Wenlong Li
DOI: https://doi.org/10.1080/03639045.2023.2212065
IF: 3.727
2023-05-20
Drug Development and Industrial Pharmacy
Abstract:The drooping process of the Xuesaitong dropping pills (XDPs) was optimized based on quality by design concept. Meanwhile, a machine vision (MV) technology was creatively introduced in this study to predict the critical quality attributes (CQAs) rapidly and accurately. This study improves the understanding of dropping process, and has reference value for the guidance of pharmaceutical process research and industrial production. The study mainly consisted of three stages: the first stage involved the prediction model to establish and evaluate the CQAs, and the second stage involved assessing the quantitative relationships between critical process parameters (CPPs) and CQAs by the mathematical models that were established through the Box–Behnken experimental design. Finally, a probability-based design space for the dropping process was calculated and verified according to the qualification criteria of each quality attribute. The results show that the prediction accuracy of the random forest (RF) model was high and met the analysis requirements, and the CQAs of dropping pills can meet the standard by running in the design space. The MV technology developed in this study can be applied to the optimization process of the XDPs. In addition, the operation in the design space can not only ensure the quality of XDPs to meet the criteria, but also help to improve the consistency of XDPs. GRAPHICAL
pharmacology & pharmacy,chemistry, medicinal
What problem does this paper attempt to address?