| Background:Data mining technology is the computer application and analysis of data management technology with the combination of a technology,which can realize accurate analysis from a large number of research information and hidden characteristics of its is difficult to detect,now the technology has been more and more into the field of traditional Chinese medicine,because the patients in the prescription of TCM symptoms vary,different classification,various types of drugs involved,according to the traditional way of study,it is difficult to from a huge number of solutions are found in the rule and correlation,the improvement of diagnosis and treatment of disease has certain limitations,and now you can through the data mining technology will be a large number of prescription summary analysis,digging out the hidden characteristics,regularities and relevance,combining TCM symptoms and experience,to improve the diagnosis and treatment scheme,can be from a new Angle to analyze the theory and experience,and interpretation of traditional Chinese medicine diagnosis and treatment to improve the patient's diagnosis and treatment effect,benefit the patients.Lung cancer on the basis of an analysis of the TCM etiology is human body qi and blood stasis,phlegm hot in lung wet poison and so on,the,in turn,the cough,bleeding,associated symptoms,including chest pain.Lung cancer is usually on the doctor of traditional Chinese medicine dialectical type belongs to evidence on the overall treatment,but because of the integrity of the treatment theory of traditional Chinese medicine and multiple targets,inaccuracy,not quantitative,make it difficult to realize can forecast precision like western medicine treatment,therefore,by means of modern advanced data mining and analysis,a summary analysis of traditional Chinese medicine therapy,to find the treatment of commonness and individuality,analysis,to improve the treatment of traditional Chinese medicine,to raising the level of clinical diagnosis and treatment of traditional Chinese medicine has a very important significance.Objective:Analysis by applying data mining technology in recent years,our hospital diagnosis and treatment of lung cancer prescription,drug designed to find out the characteristics,formula and dig out.of New Zealand law,and select the new prescription of the effectiveness of the new prescription through animal models of lung cancer and tumor suppression rate for validation,for Chinese medicine clinical diagnosis and treatment of lung cancer provide reasonable theory basis and data support.Research methods:The first part1.case collection:This study used a retrospective research method,all the cases were from December 2011-December 2014,our hospital oncology and regions of diagnosis and treatment of patients with lung cancer,collected 98 patients,121 prescription.2.patients with inclusion criteria:(1)patients with personal information,disease information,complete prescribing information records.(2)conforms to the western medicine diagnostic criteria,lung cancer patients with symptoms,chest X-ray,CT,MIR,pathological examination were diagnosed with lung cancer.(3)in patients with symptoms and signs and pulse condition,tongue conforms to lung cancer,such as,the dialectical TCM diagnostic criteria.(4)the history of treatment(including surgery,radiation and chemotherapy,and drug use,etc.)information completely.(5)the number of clinic visits to more than three times,after treatment,has the obvious symptoms or examination showed improvement.3.patients exclusion criteria:(1)clinical diagnosis is not clear,such as unknown clinical stage or pathologic types are not clear.(2)patients with other malignant tumor metastasis to the lungs.Past history(3)a merger or other malignant tumor patients.4..prescription inclusion criteria:(1)to be included in the formula should be is designed for lung syndromes;(2)the treatment with traditional Chinese medicine is given priority to,including postoperative cooperate with other support therapy,such as during chemotherapy,radiation therapy,cooperate with drug use;(3)traditional Chinese medicine compound ingredients must be completely,dose accurately,the attending clear,curative effect is true;(4)because of the symptoms,and add and subtract,within the statistical categories;5.prescription exclusion criteria:(1)the ingredients is not complete,well no medicine or repeat prescriptions;(2)the curative effect is not exact;(3)no syndrome;(4)no doses or dose not accurate;6.Data entry and database established:By excel,inheritance of Traditional Chinese medicine auxiliary platform(Traditional Chinese medicine inheritance auxiliary platform,TCMISS)(V2.5)(the software provided by the China institute of Traditional Chinese medicine academy of sciences)and other software will be collected by the patient's name,sex,age,date of visits,disease diagnosis and classification,number of TCM syndrome differentiation,prescription and medication information such as the unified field after entry,respectively.Data input of standard reference:drug name reference to"the Chinese materia medica" and "Chinese materia medica";TCM symptoms and diagnostic reference 2008 edition of the tumor diagnosis and treatment of traditional Chinese medicine guide.Establish a database,and by two audit data respectively,professional and technical personnel guarantee the accuracy of the input information is correct.7.data analysisThrough "TCMISS V2.5 system" in the "data analysis" function,through its"prescription analysis" function in the module,to analyze drug laws:(1)frequency statistics:the frequency of drugs in prescriptions from high to low order;(2)sets of rules:the support set to 40%(96 at least 38)in the prescription,the confidence level of 0.95,formula rule analysis;(3)New Zealand analysis:a set of 8 relevancy,punishment degree is set to 2,clustering analysis,and then click the "extract combination" button to find new formula,data mining,frequency analysis of drugs,to export to Excel file.The second part1.lung cancer model is set up:Take off 5 only Lewis lung cancer a tumor-burdened mouse,neck to death,from tumor,complete the cut into pieces,according to the tumor tissue 1,sterile saline mix proportion of 3,ground into a mixed suspension,after coarse filter centrifuge,remove the supernatant,residual contents after trypan blue staining,adjust the concentration of living cells to 2 ×106/mL,both sides of the vaccination to C57BL/6 j mice forelimbs axillary,0.2 mL/only.2.the preparation of traditional Chinese medicine medicinal broth:Traditional Chinese medicine formula:radix astragali 30 g,north and south straight ladybell each 15 g,15 g day wheat,radix asparagi 15 g,white mulberry root-bark 10 g,15 g of radix polygoni multiflori,15 g,Chinese wolfberry raw oysters(Fried)first 30 g,zhejiang bei 10 g;Lung cancer new reagent 1:10 g white mulberry,radix astragali 30 g,schisandra,15 g,15 g of pinellia,30 g unibract fritillary bulb,loquat leaf 15 g,madder root,10 g notoginseng 15 g,15 g yam,salvia miltiorrhiza 15 g,herba agrimoniae 10 g;The above formula of medicinal materials with distilled water for 1 h,fire boil after 5 min,turn a small fire boiled water for 30 min,poured out medicine juice and filtering,the second time to join the distilled water,fire fire boil after 5 min,turn a small fire boiled water for 30 min,to the medicine juice and filtering,mix juice and pour out for the first time,made concentration is 3.25 g/ml of liquids.3.animal group and drug intervention:Will be 30 C57BL/6 j mice inoculated tumour cells,in accordance with the block randomization method is divided into model group,the conventional formula group,the new formula group,10 in each group.Routine formula group:within 5 days after the building began to once a day in accordance with the standards of 0.2 ml/20 g weight with the conventional formula concentrated solution to mice stomach,12 days in a row.New reagent groups:within 5 days after the building began to once a day in accordance with the standards of 0.2 ml/20 g weight with concentrated new reagent solution to mice stomach,12 days in a row.Model group:within 5 days after the building began to once a day in accordance with the standards of 0.2 ml/20 g weight with sterile saline lavage to mice,for 12 days.4.materials and specimen preparation:A day after the last delivery,cervical dislocation execution groups of mice,stripping out the tumor tissue,weigh and tumors after weight,tumor inhibition rate calculation,will organize take half into the fore Ma Linzhong is fixed,the other half into the liquid nitrogen and set aside.Inhibition rate(IR)=(tumor model group,the average tumor weight-group the average tumor weight)/tumor model group the average tumor weight × 100%5.pathological dyeing:Tumors had tissue embedding into wax pieces and cut into 5 um thick slices,each take three slices of HE staining,optics microscope.6.the immune histochemical method in the detection of tumor tissue Ezrin protein,the expression of HSP70 protein and PRDX1 protein:Groups take three slices of immunohistochemical experiment,observed under optical microscope(400 x),with Biosens Digital Imaging System v1.6 measurement observation area value integral optical density(IOD)positive cells.7.reverse transcription polymerase chain reaction method(reverse transcription polymerase chain reaction,RT-PCR)in detection of tumor tissue EzrinmRNA and HSP70mRNA PRDX1mRNA expression:Will be put into liquid nitrogen trituration cryopreserved tubes in the tumor tissue,Trizol extracting total RNA,reverse transcription reaction.Take 5 ulpcr amplification products with agarose gel electrophoresis,gray scanning analysis,gene EzrinmRNA EzrinmRNA and beta actin is the expression quantity of PCR bands of the ratio of the grey value,the expression of HSP70mRNA is HSP70mRNA with beta actin PCR bands ratio of the grey value,PRDX1mRNA PRDX1mRNA and beta actin is the expression quantity of PCR bands of grey value.8.data processing and statistical methods:With SPSS20.0 statistical analysis,measurement data with mean ±standard deviation(x± s)said.Difference between groups using single factor analysis of variance method,statistics,first to each detection index data of normality test and f test;Diverge from variance analysis,further the LSD-t method are compared,and two P less than 0.05 is statistically significant.Results:1.After into screening and exclusion criteria,participants included 64 cases of patients,prescription 96,age<40 cases of patients accounted for 3.13%,the proportion of 40-50 cases patients was 12.5%,the proportion of 50-60 cases patients was 39.06%,the proportion of patients with 60-70-year-old cases(32.81%),70-80-year-old patient cases accounted for 9.37%,>80 cases patients with the proportion is 3.13%,the highest proportion of 50-60 years old age group,accounting for more than the minimum for<and>40 years old of a 80-year-old age.2.64 patients in stage 0 in 6 cases,accounting for 9.37%,? period of 19 cases of patients,29.69%,? period of 28 cases of patients,43.75%,? period in 10 cases,the patient 15.63%,? period ? case of patients,1.56%,the highest proportion of? period patients.3.64 patients in 5 cases,7.81%,lung spleen deficiency syndrome in 8 cases,accounted for 12.50%,phlegm heat resistance pulmonary syndrome in 10 cases,15.63%of qi and Yin deficiency syndrome in 11 cases,accounting for 17.18%,lung Yin deficiency syndrome,6 cases with 9.37%,deficiency of phlegm dampness syndromes in 9 cases,14.06%,Yin deficiency and internal in 8 cases,accounted for 12.50%,toxic heat flourishing in 7 cases,10.75%,the highest proportion of qi and Yin deficiency syndrome.4.96 there're a total of 121 kinds of prescription drugs,including a total of 21,frequency of 30 or more drugs salvia miltiorrhiza 89 times,88 times,codonopsis pilosula astragalus 88 times,angelica,86 times,poria cocos,84 times,81 times limit,and?77 times,77 times,snake tongue grass ephedra 75 times,root of herbaceous peony 73 times,73 times of the bark of eucommia,pinellia 71 times,fructus schisandrae 67 times,with 62 times,radix pseudostellariae 58 times,55 times unibract fritillary bulb,yanhusuo 53 times,almond 51 times,rhizoma zedoariae,48,dwarf lilyturf,47,white mulberry root-bark 44,use frequency of the first three of salvia miltiorrhiza,dangshen,astragalus membranaceus.5.The "support" is set to 40%,"confidence" is set to 0.95,the occurrence frequency of more than 38 times drug combinations have 18 on,contains 8 core drugs and salvia miltiorrhiza respectively,astragalus,atractylodes,angelica,twotooth achyranthes root,pinellia,ephedra,unibract fritillary bulb.Danshen-74 times limit,and dangshen,astragalus membranaceus,72 times,astragalus membranaceus,poria cocos,68 times,angelica,astragalus,66 times,poria cocos-ephedra 63 times,extracted by angelica 62 times,twotooth achyranthes root,salvia miltiorrhiza,62 times,pinellia tuber-astragalus membranaceus 60 times,ephedra-57 times,atractylodes unibract fritillary bulb-astragalus membranaceus 54 times,salvia miltiorrhiza-?53 times,pinellia tuber-51 times unibract fritillary bulb,fructus schisandrae,angelica 51 times,pinellia tuber-49 times,angelica radix pseudostellariae,salvia miltiorrhiza,47 of unibract fritillary bulb-ephedra 44,yanhusuo,astragalus membranaceus 42 times,extracted by ephedra 41 times.6.According to the number of prescriptions,through a combination of different parameters are extracted data pre-reading,set up the correlation of 8,punishment degree for 2,clustering analysis,get treatment prescriptions of TCM in 121 two correlation between the two drugs,the correlation coefficient of the top 15 of the medicine for poria cocos-astragalus membranaceus,ephedra,astragalus membranaceus,twotooth achyranthes root,salvia miltiorrhiza,angelica,atractylodes,poria cocos,dangshen,astragalus root,Chinese angelica,astragalus,atractylodes,pinellia,atractylodes,ephedra,salvia miltiorrhiza,twotooth achyranthes root-astragalus membranaceus and salvia miltiorrhiza-unibract fritillary bulb,angelica-unibract fritillary bulb,ephedra,pinellia,angelica root,pinellia,extracted by fructus schisandrae.7.With improved mutual information method,on the basis of the analysis results according to the correlation coefficient and punish coefficient constraints,based on the complex system entropy clustering,evolved the core 4-7 herbs combination,a total of 15,poria cocos-astragalus-h.cordata-chicken's gizzard-membrane,ephedra-radix bupleuri-dried tangerine or orange peel-astragalus membranaceus,almond-suberect spatholobus stem-achyranthes-salvia miltiorrhiza,pinellia tuber-caulis dendrobii angelica-atractylodes,notoginseng,herba agrimoniae,poria cocos-dangshen-selfheal,keel-rhizoma zedoariae-astragalus,angelica-three edges,astragalus,atractylodes-asters-caulis dendrobii-pinellia,frankincense-betel nuts-pinellia rhizoma zedoariae-system-atractylodes,ephedra-asters-coltsfoot flower-these-salvia miltiorrhiza,twotooth achyranthes root-snake tongue grass-these-it-magnolia bark-astragalus membranaceus,raw oysters-born keel-three arrises-rhizoma zedoariae-salvia miltiorrhiza-unibract fritillary bulb,rhizoma zedoariae-three arrises-asters-coltsfoot flower-snake tongue grass-angelica-unibract fritillary bulb,magnolia bark-it-astragalus angelica-white mulberry-ephedra,pinellia,yanhusuo-sweet attached-tuber fleeceflower stem-god-white mulberry-eight angelica root,pinellia,astragalus,bupleurum,Dan skin-dwarf lilyturf-day winter-extracted by fructus schisandrae.8.On the basis of the above core combination extraction,application system integration "clustering" "extract" combination of functions by the entropy level of unsupervised clustering algorithm,get three new reagent,formula 1:white mulberry,radix astragali,fructus schisandrae,pinellia,unibract fritillary bulb,loquat leaf,madder root,notoginseng,Chinese yam,salvia miltiorrhiza,herba agrimoniae;Formula 2:astragalus root,white mulberry root-bark,almonds,pinellia,white mustard,coix seed,poria cocos,incense,dried tangerine or orange peel,radix paeoniae alba,rhizoma corydalis,twotooth achyranthes root,eucommia bark,rhizoma sparganii,rhizoma zedoariae,mulberry parasitism,centipede,scorpion;Formula 3:radix astragali,fructus schisandrae,radix pseudostellariae,coltsfoot flower,aster,unibract fritillary bulb,radix peucedani,loquat leaf,almond,bitter orange,snake tongue grass,barbed skullcap herb,rhizoma zedoariae,triangular,salvia miltiorrhiza,keel,raw oysters,turtle shell.9.Experiment results show that the model group(6.75 ? 1.04 mm)the average tumor weight g,higher than the conventional formula and new reagent group(P<0.05),the conventional formula group of the average tumor weight was(5.86 ±0.86)g,higher than the new reagent group(the average tumor weight was 4.95 ±0.62 g)(P<0.05);Conventional formula group of the tumor inhibition rate was 13.18%,the new formula group of the tumor inhibition rate was 26.67%,higher than that of conventional formula group(P<0.05).10.The distribution in the tumor cells in model group section view,less necrosis area;Conventional formula group slice view tumor cell density was lower than those of model group,the necrosis area increased;New reagent group slice view tumor cells relative to a minimum,necrotic area the most.11.Under the microscope view,the model groups view visible in a large number of Ezrin positive cells,and Ezrin IOD value is(12.41 ± 2.77),the expression level is higher than the other two groups(P<0.05);Routine formula group view visible in a small amount of Ezrin positive cells,Ezrin IOD values for(8.36 ± 1.25),the expression level is higher than the new reagent group(IOD value was 6.03±1.04)(P<0.05);Stripe color the brightest in addition,the model group,EzrinmRNA transcription for(14.28 ±2.76),are higher than the conventional formula and new reagent group(P<0.05),the conventional formula of stripe color slightly dim,EzrinmRNA transcription for(10.25 ±2.11),were higher than in new reagent group(P<0.05),a new reagent group stripe color dim,EzrinmRNA transcription for(7.30±1.62),lower than the other two groups(P<0.05).12.Under the microscope view,the model groups view visible in a large number of HSP70 positive cells,and HSP70 IOD value is(1.86 ± 0.33),the expression level is higher than the other two groups(P<0.05);Conventional view visible in a small amount of HSP70 masculine cell formula group,the IOD value of HSP70(1.24±0.27),the expression level is higher than the new reagent group(IOD value was 0.52 ±0.21)(P<0.05);Stripe color the brightest in addition,the model group,HSP70mRNA transcription for(6.88 ± 1.26),are higher than the conventional formula and new reagent group(P<0.05),the conventional formula of stripe color slightly dim and HSP70mRNA transcription for(4.37 ±0.94),were higher than in new reagent group(P<0.05),a new reagent group stripe color dim,HSP70mRNA transcription for(1.95 ±0.42),lower than the other two groups(P<0.05).13.Under the microscope view,the model groups view visible in a large number of PRDX1 positive cells,and PRDX1 IOD value is(0.88 ± 0.05),the expression level is higher than the other two groups(P<0.05);Conventional formula group view a small PRDX1 positive cells,PRDX1 IOD values for(0.61 ±0.03),the expression level is higher than the new reagent group(IOD value was 0.37 ± 0.02)(P<0.05);Model group stripe color is most bright,PRDX1mRNA transcription for(9.30±2.16),are higher than the conventional formula and new reagent group(P<0.05),the conventional formula of stripe color slightly dim,PRDX1mRNA transcription for(7.48 ±1.92),were higher than in new reagent group(P<0.05),a new reagent group stripe color dim,PRDX1mRNA transcription for(5.02 ± 1.33),lower than the other two groups(P<0.05).Conclusion:1.The data mining technology can dig out the formula of the lung cancer drug characteristics,formula.2.The new formula 3 is obtained by data mining make a diagnosis and give treatment of lung cancer.3.The new formula 1 can reduce the tumor weight and raise the inhibition rate of lung cancer in mice.4.The new formula 1 can reduce tumor of lung cancer in mice tumor tissues Ezrin,HSP70,PRDX1 protein and gene expression of the tumor inhibition may be related to this. |
PDF Full Text Request