Professor Lv Renhe Treatment Of Diabetic Academic Thought And Its Inheritance Method

Professor Lu Renhe, as a physician-in-charge and a postgraduate candidate teacher, has devoted to the research of traditional Chinese internal medicine for over 50 years, especially in the research of diabetes. He has collected a lot of experiences concerning the prevention and treatment of diabetes. How to summarize and inherit the experiences of such experts of Traditional Chinese Medicine (TCM), is the most key problem in the study of TCM. But, from the analysis of present inheritance situation of TCM, the valuable results are still limited. By finding a method, which can be used to analysis complex system, and which is good at the analysis of connection, we may bring the new direction to the inheritance of traditional Chinese medicine.Comparing the basic thinking mode of modern natural sciences and TCM, we find out that, they share some same thinking modes. So, from a methodological point of view, there is a possibility to using the new technology and practical method of modern natural sciences in the research of TCM.Therefore, based on the data warehousing technology, by the application of data mining technology and the support machine vector (SVM) method, we try to clear up and excavate the experiences of professor Lu. The basic research steps are:"The data acquisition", refers to the process of gathering recipes prescribed by professor Lu in treating diabetes. In this stage, in order to maintain the Professor Lu's originally diagnosis characteristic, the key point is not to bother the diagnosis process of the professor;"The information storage", refers to the preservation of diabetes medical notes. In this research, we use"the standardization and the uniformization"method, namely"the full text analysis"and"the subject index", in processing of diabetes medical notes. All these are based on the platform of"fundamental data-base system of TCM"developed by Beijing University of Chinese medicine. The knowledge discovering, refers to the analysis of diabetes medical records. By using the data-base system, we manage to fulfill the knowledge discovery of the prescriptions prescribed by professor Lu, which also are the academic thoughts and the speculation rules of professor Lu.We also use"the Chinese medicine prescription intelligence analysis system"(CPIAS, also developed by Beijing University of Chinese medicine) and the SVM method to excavate the experiences of professor Lu Renhe.Result: In the medical notes database, based on the frequency statistics and the analysis of Chinese medicine dialectical essential factors which are contained in the typical medical records, we found some medication characteristic and the dialectical mentality of professor Lu: the herb which is most frequently used by professor Lu in the prescriptions of treating diabetic patient, is Radix SaLuiae Miltiorrhizae (Dan Shen );the main concerning is to prompt blood circulation by removing stasis, and to dredge meridians and activating collaterals; at the same time, activating qi is also important, especially harmonizing the liver qi; and we also found that, Radix Achyranthis Bidentatae (Niu Xi ), Rhizoma Cibotii (Gou Ji), and Radix Dipsaci (Xu Duan) are composed of a unique herb pair, which is used frequently by professor Lu. The conclusions also can be explained well by professor Lu's basic academic thought of treating diabetes which has been summarized formerly. And the statistical results show that, the academic thought summarized formerly in treating diabetes of professor Lu, are highly consistent with clinical practice.According to professor Lu's division of the periods in diabetes, we divide 128 typical medical records into"PI DAN","XIAO KE","XIAO DAN"3 categories. By separately analyzing the Chinese medicine dialectical essential factors which are contained in the typical medical records in each category, we found that: the statistical results drew from the analysis of each category of the diabetic records, invoLuing the analysis of symptoms and signs, the location of disease, the cause of the disease, the pathological mechanism of disease, the nature of disease, and the differentiation of the disease, are highly conformed with the professor Lu's division theory of diabetes, which has been formerly summarized.By using the multi-level, multi-essential factors, and multi-relational inquiry function of Chinese medical records database, we found out the ways of professor Lu in treating the diabetic patients with"sore and painful waist"symptom. Invigorating the kidney is essential, dredging meridians and activating collaterals are also very important. All the results obtained also indicated that, the medical records database can satisfy the request of correlation analysis of traditional Chinese medicine, therefore, would be helpful in the knowledge excavation of Chinese medicine.Through the introduction of"the Chinese medicine prescription intelligence analysis system"(CPIAS), we made the quantification analysis of a specific formula chosen from a typical diabetic case. The analysis concludes the quantitative calculation of the relation between dosage of medicine and action of formula, the relation between action of Chinese material medicine and action of formula, the relation between syndrome and action of formula, and the relation between symptoms and the action of formula. The system analysis showed: the most distinguished synthesis flavor of the formula is the flavor of bitter, which indicates the function of heat-clearing and fire-purging; the analysis of main meridian tropism of the formula indicates that the liver is the most important organ to be harmonized in treating diabetes; and the high attention is paid to prompt blood circulation by removing stasis by professor Lu in the process of treating diabetes.SVM (Support Vector Machine) is a useful technique for data classification and the underlying theory of SVM has a solid mathematic foundation. A classification task usually involves with training and testing data which consist of some data instances. Each instance in the training set contains one"target value"(class labels) and several"attributes"(features). The goal of SVM is to produce a model which predicts target value of data instances in the testing set which are given only the attributes. The research carried on by the software of CMSVM2.0. A kind of Chinese formulae which all have the specific effect of activating blood flow and removing blood stasis was set as the training data. The model produced was intended to distinguish the professor Lu's prescriptions which also have the effect of activating blood flow and removing blood stasis from all kinds of prescriptions. Results showed that: the capacity of the model was not good enough to fulfill the target. The reasons may be: the number of the training sample is too limited and the"target value"of the training formulae is not distinctive enough. As a first attempt to the application of SVM in the Chinese formulae classification, this research provides an operating procedure which may be helpful for the further study in the future.