| In southern China, the weather is hot, damp and rainy. Febrile diseases (known in Chinese "Wen" disease) commonly seen here as exogenous affection are mostly damp-heat syndrome. The pathological change is around spleen and stomach. The damp-heat syndrome of febrile disease includes damp-wen, summer-damp, damp-heat epidemic and summer-wen with dampness, etc. Clinical manifestation is varied, symptoms complicated, not easily differentiated. The damp-heat syndrome of febrile diseases can be associated with different viscera such as lung, heart, spleen, small intestine, large intestine, spleen, liver, kidney etc. The symptoms are varied. The main symptoms and associated symptoms have not been well differentiated in literature. Since differentiation of syndromes is the basis for treatment, without proper regulation on the syndrome differentiation, clinical research and international exchange cannot be carried out at a deep-rooted level. Research is being carried out in this direction.Objective1. Through literature research, screen out the main symptoms and the basic fever types.2. Regulate the main symptoms of damp-heat syndrome associated with different viscera.3. Regulate the degree of damp and heat in the differentiation of damp-heat syndrome of febrile disease. Research Significance1. Advantageous to the regulation of the differentiation of damp-heat syndrome of febrile diseases.2. Facilitate damp-heat syndrome research, regulation of the main symptoms supplies clinical data for statistics, analysis and result determination giving rise to a great research significance.Methodology1. Carry out statistical analysis on the frequency of symptoms appearing in damp-heat syndrome as recorded in Chinese medicine literature and neoteric clinical literature. Screen out the fever types, main symptoms and clinical manifestations.2. Sort out and regulate the main symptoms and associated symptoms due to damp-heat syndrome in different viscera, based on ancient Chinese medicine literature and neoteric clinical literature.3. In accordance with the four techniques of diagnosis, the eight principal syndromes and the theory of viscera and their manifestations, sort out and group data in literature and ancient cases. Discretise them into computer program to analysis the significance of each symptom and establish the standard of the syndromes. Thereafter use the results to test damp-heat syndrome cases in Chinese Medicine textbooks.Tools1. Collect 99 documented cases and 105 neoteric clinical cases of damp-heat syndrome of febrile disease, making a total of 204 cases for analysis. 85 cases were taken from textbooks for testing. This research used neural network to establish models.2. The beauty of neural network is that it can learn by itself the intrinsic rule of the data and therefore does not require to preset the relationship of the elements to be studied. Neural network has been successfully applied in the health field including such applications as medical diagnosis, medical imaging analysis, biomedical analysis and drug development. Results1. Apart from "body temperature does not fly high on the surface" as the general fever character of damp-heat syndrome of febrile disease from the statistical results, the basic fever types show in2. Gist of differentiation of syndromes: fever and tongue plus one or two syndromes associated with the heat evil and one or two syndromes associated with the damp evil will establish the diagnosis. 3. The primary symptoms of damp-heat diseases at different viscera 4. Gist of differentiation of syndromes: The common syndromes and tongue plus one or two syndromes associated with the heat evil or one or two syndromes associated with the damp evil will establish the diagnosis. Conclusion1. Statistical analysis shows that fever is the main symptom of the damp-heat syndrome of febrile disease. Neural network indicates that dull fever, or high fever or no abatement of fever after sweating are the main fever types. Fever has a correlation with sweating. Fever has a correlation of accessory symptom with aversion to cold.2. Statistical results and neural network analysis both indicate that the main symptoms of damp-heat syndrome are fever, thirsty and chest distress.3. Statistical results and neural network analysis both indicate that greasy coating of tongue and distension of stomach and abdomen are also chief manifestations of damp-heat syndrome.4. The frequency of "aversion to cold" appearing in literature is 26.3%, indicating that "aversion to cold" can be a symptom of damp-heat syndrome.5. The frequency of "poor appetite" appearing in literature is 12.5%, indicating that while "poor appetite" may be a symptom of damp-heat syndrome, but not really necessary.6. The weighting results of neural network indicate that1) For the properties of damp more than heat: The pathogenic cause: Damp bound in spleen, Qi mechanism stagnantThe common symptoms: Dull fever and scanty sweat, Sticky throat without polydipsia or thirst but not drink, heaviness of head and body;2) For the properties of damp about equal to heat: The pathogenic cause: Damp and heat obscure spleen and stomachThe common symptoms: No abatement of fever after sweating, dottiness but not be sick of, Bitter and/or greasy taste, thirst not drink much;3) For the properties of heat more than dampness: The pathogenic cause: Heat act on stomach, damp in spleen cannot be clearedThe common symptoms: high fever, bitter taste and symptoms of polydipsia, abnormal stools and urine.7. Poor correlation of the testing samples indicate the differentiation of damp-heat syndromes in textbooks is not clear in respect of symptoms and pathological changes, limited in the scope of discussion and clinical manifestation. For example the problem of summer heat with damp is dominant.8. This research shows that neural network can be used in the classification TCM syndromes, believes to be the first test in the "Science of Epidemic Febrile Diseases" of Chinese Medicine. 9. There appears to have insufficient number of samples for analysis. Further research can be directed to increasing the number of samples, exploring the various parameters affecting the optimization of the neural network, hopefully to obtain even better results. |
