人工智慧於藥物崩散製程最佳化之研究

The Parameters Design for Disintegrate production using Artificial Intelligence

鄭蔚泓(Wei-Hong Cheng)

指導教授 ： 邱昭彰

元智大學/資訊學院/資訊管理學系/碩士(2007年)

摘要

一般產業對於生產條件的控制，通常都是根據其經驗的判斷或是以試誤法進行製造，往往都是費時費力又消耗許多資源。藥物崩散劑會影響藥物的崩散、溶離及吸收，對於藥物的藥效影響很大。本研究的目的是要找出最適當崩散劑的製程條件，利用類神經網路預測崩散劑製程的條件因子與沉降體積的關係，再根據預測結果以基因演算法依照所需的條件找出最佳製程，製造出各種不同等級的崩散劑，並且可以根據生產的需求限制製程的控制因子。其實驗結果顯示利用人工智慧所計算出的生產條件，與實際生產所得到的崩散劑兩著的沉降體積兩者比較，其誤差非常小，因此可利用此一方法改善崩散劑的製程。

關鍵字

崩散劑 ； 類神經網路 ； 基因演算法

並列摘要

In general, to control the working conditions under production will adopt trial and error method or simply judge by the existing experience, which is quite time or labor consuming and even wasting plenty of resources. Disintegrate will affect the disintegration, dissolution and abstraction of the medicine and have significant effect to the medicinal virtues of it. The main purpose of this study is trying to find out the best production condition of Disintegrate. By using artificial neural network firstly to observe the relationship between the productive factors and its setting volume under production process of croscarmellose sodium. Based on the above production result then using Genetic Algorithm to conclude the best production process so as to make Disintegrate in different grade in the mean time to restrict the factors during production as per production requirement.
As according to the report from laboratory, comparing the setting volume of those precisely analyzed by artificial intelligence and those actually obtained from the production, the deviation is quite small and negligible. Consequently, the above method is good enough to improve the production process of Disintegrate.

並列關鍵字

Disintegrate ； artificial neural network ； Genetic Algorithm

參考文獻

1.Ferrero, C., Munoz, N., Velasco, M. V., Munoz-Ruiz A., and Jiménez, C., “Disintegrating Efficiency of Croscarmellose Sodium in A Direct Compression Formulation,” International Journal of Pharmaceutics, Vol. 147, pp. 11-21, 1997.

2.Rockville, MD, United States Pharmacopeia and National Formulary 23. United States Pharmacopeia, pp.2993-2994, 1995.

4.Umang, S. and Larry, A., “Evaluation of the Functional Equivalence of Crospovidone NF from Different Sources. II. Standard Performance Test,” Pharmaceutical Development and Technology, Vol. 6, Issue3, pp. 419-430, 2001.

6.Rocksloh, K., Rapp, F.R., Abed, S., Muller, W., Reher, M., Gauglitz, G., and Schmidt, P. C., “Optimization of Crushing Strength and Disintegration Time of A High-Dose Plant Extract Tablet by Neural Networks,” Drug Development and Industrial Pharmacy, Vol. 25, Issue 9, pp. 1015-1025, 1999.

7.Joseph, V., Desmond, J. and David, J., “Pharmacokinetic Parameter Prediction from Drug Structure Using Artificial Neural Networks,” International Journal of Pharmaceutics, Vol. 270, pp. 209-219, 2004.