| Chinese Patent Medicine (CPM), taking Chinese herbal medicine as raw material, is the essence of effective prescription according to the medical experts of our country by medical practice to create and sum up for hundreds of years. The quality control of CPM has become the key to the development of modernization and internationalization. The raw material quality and production process will directly affect the quality and curative effect of the finished product. Infrared spectrum with the "fingerprint" characteristics has become one of the most effective means of identification for CPM. The same CPMs from different manufacturers are different because of the raw material and the production process and so on. Firstly, the infrared spectrum fingerprints of twenty kinds of CPM were analyzed and investigateed applied different methods of chemical metrology and the average infrared spectroscopy fingerprint library of each CPM was founded in this paper. Lastly, each fingerprint of CPM has been searched in the whole spectrum range.Chapter one:Introduction. At first, this article summarizes the twenty kinds of CPM. Then it surveys the documents about the application of clustering analysis and Artificial Neural Network (ANN) in CPM infrared spectrum. Finally, this paper expounds the research purpose, significance and prospect. The 66 references were cited.Chapter two:The optimization of experimental conditions. The experimental conditions were optomized using single factor method for the investigation of CPM infrared spectrum with solid tablet and the optimum experimental conditions are as follows: particle size of 360 mesh, mixing ratio with KBr of 6:100, quality of 95 milligram, tablet pressing pressure of 70kN and tablet pressing time of 10s.Chapter three:Study on the infrared fingerprints of clearing heat medicines. The infrared spectrograms of 4 kinds of CPM those belong to clearing heat medicine samples were determined, respectively. The consistency of the infrared spectrograms of the same CPM was investigated by cluster analysis and the results show that the infrared spectrograms of samples from the same manufacture and different manufactures were in good agreement. The ANN was applied to classify the 4 kinds of clearing heat medicines and the results show that the identification of 4 kinds of CPM can be achieved by ANN. The infrared fingerprints of 4 kinds of clearing heat medicines were established by average method.Chapter four:Study on the infrared fingerprints of tonic class medicines. The infrared spectrograms of 4 kinds of CPMs those belong to tonic class medicine samples were determined, respectively. The consistency of the infrared spectrograms of the same CPM was investigated by cluster analysis and the results show that the infrared spectrograms of samples from the same manufacture and different manufactures were in good agreement. The ANN was applied to classify the 4 kinds of tonic class medicines and the results show that the identification of 4 kinds of CPM can be achieved by ANN. The infrared fingerprints of 4 kinds of tonic class medicines were established by average method.Chapter five:Study on the infrared fingerprints of regulating vital energy medicines. The infrared spectrograms of 5 kinds of CPM those belong to regulating vital energy medicine samples were determined, respectively. The consistency of the infrared spectrograms of the same CPM was investigated by cluster analysis and the results show that the infrared spectrograms of samples from the same manufacture and different manufactures were in good agreement. The ANN was applied to classify the 5 kinds of regulating vital energy medicines and the results show that the identification of 5 kinds of CPM can be achieved by ANN. The infrared fingerprints of 5 kinds of regulating vital energy medicines were established by average method.Chapter six:Study on the infrared fingerprints of cure indigestion medicines. The infrared spectrograms of 4 kinds of CPM those belong to cure indigestion medicine samples were determined, respectively. The consistency of the infrared spectrograms of the same CPM was investigated by cluster analysis and the results show that the infrared spectrograms of samples from the same manufacture and different manufactures were in good agreement. The ANN was applied to classify the 4 kinds of cure indigestion medicines and the results show that the identification of 4 kinds of CPM can be achieved by ANN. The infrared fingerprints of 4 kinds of cure indigestion medicines were established by average method.Chapter seven:Study on the infrared fingerprints of regulating blood and arresting leucorrhoea medicines. The infrared spectrograms of 3 kinds of CPM those belong to regulating blood and arresting leucorrhoea medicine samples were determined, respectively. The consistency of the infrared spectrograms of the same CPM was investigated by cluster analysis and the results show that the infrared spectrograms of samples from the same manufacture and different manufactures were in good agreement. The ANN was applied to classify the 3 kinds of regulating blood and arresting leucorrhoea medicines and the results show that the identification of 3 kinds of CPM can be achieved by ANN. The infrared fingerprints of 3 kinds of regulating blood and arresting leucorrhoea medicines were established by average method.Chapter eight:The search of infrared fingerprint library of CPM. The infrared fingerprint library of CPM was established using OMNIC data processing software. The infrared fingerprints of 43 kinds of CPM were recorded as a standard pattern in the spectral database. The whole spectrum correlation retrieval was carried out using OMNIC software spectrum search function for the twenty kinds of CPM within 258 samples in the paper. The results show that the infrared spectrum of 228 CPM samples is matching with the infrared standard fingerprint, respectively. The matching accuracy of 88% was obtained. The established CPM infrared fingerprint standard database has good fingerprint, characteristic and accuracy for the search of CPM. |
PDF Full Text Request