The Study On Scent Fingerprint Of Chinese Herbal Based On Bionic Olfaction

With the development of analytical techniques and desirable progress of relevant researches about Chinese herbal medicines, both the compilations of pharmaceutical attributes and the comparison of componential distributions of herbal samples have been becoming more and more important for the medicinal modernization, industrialization, internalization, in particular quality assessment, to support their uses. Adequate and rational quality assessement method is very important to meet the criteria needed to the modernization of Traditional Chinese Medecines(TCM). It is usually difficult to properly evaluate and control the quality of Chinese Materia herbal since the current quality standards of TCM only deal with the analysis of a single marker compound. In this thesis, a quality assessment method using Bionic Olfactory system to make a study of scent fingerprint of Chinese herbal was presented, and the methodology was systemtically studied. This paper proposes a research scheme trying to create olfactory scent fingerprint of Chinese herbal,with which we adopted stoichiometric analysis to extract information of producing area and medical effect of the target Chinese traditional medicines from abundant Olfactory system scent data.In consideration of the deductive demand, Scent Fingerprint of Chinsese herbal reseach based on bionic Olfactory system methodology was further studied,the mechanism of Ofactory system was descriptive systemtically, exploring methods for building the bionic Olfactory scent fingerprint of Chinese herbal ,under the guidance of Chinese medicine theories,studied the scent of herbal, designed a device which is fitted for detecting the scent of herbal. Confirm the study of hot Chinese herbal is breakthrough point to build Ofactory scent fingerprint, In our experiment, we choose four types of Chinese herbal to build scent fingerprint, as follow:the GongDong BoHe, SiChuan HongHua, GuangXi BoHe, and GongDong JuHua.In thesis, we make a mathematical analysis of Carbon Black-Polymer sensor conductive mechanism inside, and set up a mathematical function for sensors array when detecting the odour, the paper introduces the application of wavelet transform in the analysis of the E-nose signal, wavelet de-noising by threshold is a self-adaptation solution, which is based on the difference between signals and noises in the distribution characteristic under wavelet transform domain by degree of energy centralization. Especially the threshold selection presents good adaptability and the result after de-noising has fine smoothness. The simulation result shows the wavelet transform is superior and effective.In experiment, using Cyranose@320 E-nose to detect the scent of herbal, it produces 32 dimension respond data. Build the olfactory scent fingerprint about the herbal species, the result shows that the scent fingerprint can classify four basic types well in their principle component analysis. And analyzed the Chinese traditional medicine "BoHe" by the way of PLS regression, saticfactory results are obtained to prove that olfactory scent fingerprint is a promising analyzing means in the study of Chinese traditional medicines between two different producing area.Based on the bionic olfactory technology to build scent fingerprint of Chinese herbal is feasible, and have shown favorable development prospects.