Creaton of tablet database and visulalization of its internal structure

建立片剂数据库，以设计理念为依据进行片剂质量设计。阐明了活性成分对片剂抗拉强度和崩解时间等性能的影响。在研究原料药的效果之前，根据扩展单纯形搜索对片剂基础配方（安慰剂片）进行优化。优选的安慰剂片硬度足够，崩解迅速。将14种有效成分分别掺入最佳安慰剂片剂中制备模型片，考察片剂的抗张强度、崩解时间等性能。在实验数据的基础上，建立了综合广义回归分析的集成人工神经网络。建立了有效成分的理化性质与片剂性质之间的可靠关系模型。

A tablet database was created in order for designing tablet formulations based on the quality by design concept. The contribution of active ingredients (APIs) to tablet properties such as tensile strength and disintegration time was clarified. Before investigating effects of APIs, the tablet base formulation (placebo tablet) was optimized according to an expanded simplex search. The optimal placebo tablet showed sufficient hardness and rapid disintegration. Model tablets were prepared by incorporating each of 14 kinds of active ingredients into the optimal placebo tablet, and then examined the tablet properties, including tensile strength and disintegration time. Based on the experimental data, an ensemble artificial neural network incorporating general regression analysis was conducted. A reliable model of the correlation between the physicochemical properties of the active ingredients and the tablet properties was thus constructed.