ObjectiveTo explore the method of optimizing the performance of traditional Chinese medicine （TCM） syndrome diagnostic models using Stacking ensemble learning. MethodsTaking the construction of TCM syndrome diagnostic model for lung cancer as an example， 2598 cases of clinical symptoms and signs from lung cancer patients in 9 hospitals were used as independent variables （i.e.， feature variables）， TCM syndrome information as dependent variables， and the clinical data were divided into training set and testing set in 8：2 ratio according to random number table method using Python 3.7 software. The stable features of TCM syndrome of lung cancer were screened using chi-square test， Spearman's correlation test， and Least Absolute Shrinkage and Selection Operator （LASSO） logistic regression analysis； nine machine learning algorithms are trained， including support vector machines （SVMs）， k-nearest neighbors （KNN） algorithm， Random Forest （RF）， Extremely Randomized Trees， Extreme Gradient Boosting （XGBoost）， Lightweight Gradient Boosting （LightGBM）， Adaptive Boosting （AdaBoost）， Gradient Boosting （GB） and the multi-layer perceptron （MLP）， to obtain 9 basic models. Four models with better performance were screened out from the above basic models and fused to form a fusion model by using the Stacking ensemble learning， and the fusion model was trained twice by the above nine machine learning algorithms and evaluated by accuracy rate， micro-average ROC curves， area under the curve （AUC）， and confusion matrix metrics， to screen the optimal diagnostic model. ResultsAfter data processing， 79 stable features and 13 TCM syndromes were obtained. In the basic model training， the comprehensive performance of RF， ExtraTrees， MLP and SVM basic models were better， so the predicted distributions of the syndromes of these four models were used as the secondary training data， and nine fusion models were obtained based on the Stacking ensemble learning （SVM， KNN， RF， ExtraTree， XGBoost， LightGBM， GB， AdaBoost， MLP）. Among them， the XGBoost fusion model performed the best， with an accuracy of 0.850 and 0.838 in the training set and test set， respectively， an overfitting difference of 0.012， and an area under the micro-average ROC curve of 0.996. All fusion models showed an improvement in accuracy and area under the micro-average ROC curve compared with the base model in the test set. ConclusionTaking the TCM syndrome information of lung cancer as an example， the XGBoost fusion model has significant advantages in improving the diagnostic performance of TCM syndrome information of lung cancer through Stacking ensemble learning. It can be seen that the advantages of Stacking ensemble learning to integrate multiple models and effectively improve the diagnostic efficiency of TCM diagnostic models， which provided a methodological reference for similar studies.

Objective To explore the interventional mechanism of Bufei Yishen formula on chronic obstructive pulmonary disease from pathway level.Methods Macrophage inflammatory model was established by LPS stimulation.Based on gene set enrichment analysis(GSEA)method,macrophage inflammation related pathways were screened,and normalized enrichment score(NES)was used to identify pathways that were reversed by traditional Chinese medicine treatment,revealing the interventional mechanism of Bufei Yishen formula and its compatibility.Results The NES of Bufei Yishen formula was-1377.23,among which the NES of Bushen compatibility was-485.07,that of Huoxue was-351.86,Huatan was-303.71,and Yiqi was-236.59.There were 213 significantly disturbed pathways reversed after the intervention of Bufei Yishen formula,among which there were 184 reversed pathways for Huoxue compatibility,147 reversed pathways for Bushen,134 reversed pathways for Huatan,133 reversed pathways for Yiqi,and the reversal rate was 75.41%,60.25%,54.92%,54.51%,respectively.90 pathways,including TGF-β production,were significantly reversed in the four compatibilities.Positive regulation of cytokine production involved in inflammatory response etc were specifically reversed pathways for compatibility.Conclusion The intensity of Bufei Yishen formula that reversed macrophage-inflammatory related pathways was in order of Bushen,Huoxue,Huatan and Yiqi compatibility.And the number of pathways that could be reversed by the compatibility of Bufei Yishen formula was Huoxue,Bushen,Huatan and Yiqi.Bufei Yishen formula could regulate the common and specific reversal pathways of the compatibilities to intervene the inflammatory response.

Small intestinal neuroendocrine tumor is a kind of tumor arising from neuroendocrine cells and with neuroendocrine function. Because of non-specific clinical manifestations, these patients are often diagnosed at advanced stage and the treatment method is lacking. In 2019, World Health Organization updated the classification of neuroendocrine tumor. In recent years, liquid biopsy, pancreastatin, somatostatin receptor function imaging provide a new possibility for early diagnosis and monitoring prognosis. The mammalian target of rapamycin inhibitor everolimus, the tryptophan hydroxylase inhibitor telotristat ethyl and the tyrosine kinase inhibitor surufatinib bring a new breakthrough to improve the clinical symptoms and prolong the survival time of small intestinal neuroendocrine tumor. This paper reviews the current situation and the latest research progress of small intestinal neuroendocrine tumor.

Objective:To investigate the differential expression of brain-expressed X-linked (BEX) family genes in pan-cancer and its value in diagnosis and prognosis of pan-cancer.Methods:RNA sequencing (RNA-seq) data, survival data, immune subtypes, the stem cell scores based on RNA and DNA methylation of 33 different tumors from The Cancer Genome Atlas (TCGA) database were downloaded from the online database of University of California, Santa Cruz (UCSC Xena) on April 10, 2022. The limma package of R software (V.4.2.0) was used to analyze the expression of BEX family genes in the TCGA database. The differential expression of BEX family genes in pan-cancer tissues and normal tissues was compared by using Wilcox test. Pan-cancer patients were divided into high expression group and low expression group according to the median expression level of BEX family genes; Kaplan-Meier survival analysis was used to evaluate the relationship between the expression of BEX family genes and the overall survival (OS) of patients; Cox proportional risk model was used to analyze the effect of the expression of BEX family genes on OS in pan-cancer patients and then the forest map was drawn. The correlation of the expression of BEX family genes with tumor microenviroment and tumor stem cells in pan-cancer patients was analyzed based on the correlation index Cor value. Spearman correlation analysis was used to analyze the correlation between the expression of BEX family genes and tumor microenviroment and cancer stem cell index in gastric cancer tissues. The RNA-seq of different tumor cell lines and drug sensitivity data download from the CellMiner database were used to analyze the correlation between the expression of BEX family genes and drug sensitivity. The correlation of pan-cancer and gastric cancer immune subtypes with the expression of BEX family genes was analyzed by using Kruskal test.Results:BEX3 was highly expressed in pan-cancer tissues in TCGA database, BEX2 and BEX4 were moderately expressed in pan-cancer tissues, and BEX1 and BEX5 were relatively low expressed in pan-cancer tissues. The expressions of BEX2, BEX3 and BEX4 were the highest in cholangiocarcinoma, the expression of BEX5 was the highest in endometrial neoplasms, and the expression of BEX1 was the highest in invasive breast cancer. Compared with normal tissue samples, the expressions of BEX family genes were up-regulated or down-regulated in various cancers (all P < 0.05). Survival analysis showed that the expressions of BEX family genes were associated with the OS of various cancers. Some tumor patients with high expressions of BEX1, BEX3, BEX4 and BEX5 had better OS compared with those with low expressions, and the differences were statistically significant (all P < 0.05). Other patients with high expression of BEX family genes had worse OS compared with those with low expressions, and the differences were statistically significant (all P < 0.05). Cox regression analysis showed that the high expression of BEX1 for stomach neoplasms; the high expression of BEX2 for acute myeloid leukemia, thymoma and endometrial neoplasms; the expression high of BEX3 for squamous cell carcinoma of head and neck,sarcoma, stomach neoplasms and endometrial neoplasms; the high expression of BEX4 for rectal adenocarcinoma, stomach neoplasms and endometrial neoplasms; the high expression of BEX5 for renal suspicious cell carcinoma and thymoma were risk factors for OS (all P < 0.05).The expression of BEX family genes was negatively correlated with the stromal score of most cancers (all P < 0.05), and positively correlated with the stem cell score (all P < 0.05). The expression of BEX family genes was negatively correlated with cancer stem cell index of gastric cancer ( P < 0.05), and was positively correlated with matrix score and estimated total score (all P < 0.05). Among different tumor cell lines in CellMiner database, BEX family genes were closely related to drug resistance of vemurafenib (Cor = -0.368, P = 0.004), Kahalide f (Cor = -0.391, P = 0.002), O-6-benzylguanine (Cor = -0.375, P = 0.003) and other drugs. All genes in the BEX family were related to the immune subtypes of pan-cancer and were highly expressed in C5 subtype (all P < 0.05).For gastric cancer, all genes showed high expression in the C3 subtype (all P < 0.05), except BEX5 ( P = 0.24). Conclusions:The expression of BEX family genes is closely related to the prognosis of pan-cancer patients, and has an impact on the tumor microenvironment, cancer stem cells and drug sensitivity. BEX family genes may be potential biomarkers for diagnosis and prognosis of pan-cancer.

【Objective】 To study the safety, effectiveness and nursing of blood/fluid warmer during the process of plateletpheresis in winter. 【Methods】 The blood re-transfusion speed during plateletpheresis in winter and the time of blood passing through the blood return pipeline was counted. The vitro blood was heated with a blood/fluid warmer under different temperature settings, and the rising speed of blood temperature was recorded. The blood samples were tested for blood routine examination, free Hb and erythrocyte morphology at 0, 15 and 30 minutes. In the process of plateletpheresis in winter, the blood donors′ ear temperature and the skin temperature near the reinfusion needle in the observation group and the controls were measured, and the blood donors were observed for shivering, arm chills, pain or other discomfort. After the blood donation, the thermal comfort was evaluated. 【Results】 There was no difference in the results of routine blood test and plasma free Hb test of vitro blood after warming at 41℃, 42℃ and 43℃ for 0, 15 and 30 minutes (P>0.05), and no change in erythrocyte morphology was found. The skin temperature near the reinfusion needle (before vs. after the start of phlebotomy) was statistically different by applying blood/fluid warmer or not(P<0.05), and no difference in the temperature between the start and end of phlebotomy was observed in the observation group(P>0.05). The vitro blood heating experiment showed that when the room temperature was within 22~24℃, the blood retransfusion speed was (100-120) mL/min; after the application of blood/fluid warmer, the temperature of reinfusion blood could be raised from 27℃ to 33~37℃. The proportion of feeling comfortable and very comfortable and the score of thermal comfort in the blood donors who used the warmer were higher than those in the controls (P<0.05). When the temperature of the warmer was set above 38℃, the average score of thermal comfort of blood donors was above 8. 【Conclusion】 It is safe to apply the blood/fluid warmer during the plateletsapheresis in winter, which can significantly improve the comfort of blood donors and reflect the humanized service of blood stations, and is worth popularizing.

Objective To study the characteristics of a rat model of migraine with hyperactivity of liver-yang and blood stagnation,which was established by Fuzi Decoction combined with electrical stimulation of the trigeminal ganglion.Methods The 30 SD rats were divided into normal group,model group,and Zhengtian pill group(1.6 g·kg-1),with 10 rats in each group.Rats in the model group and Zhengtian pill group were fed with aconite decoction(2.00 g·kg-1)once a day for 28 day to establish the hyperthermia model;On the 15 day of intragastric administration,rats in the Zhengtai pill group were simultaneously given Zhengtai pill solution once a day for 14 day;At 29 day,the trigeminal ganglion was stimulated to create the migraine model of hyperliver and stasis.After 30 min of the last dose,the macroscopic signs and behaviors were observed,and the blood rheology was detected by blood visticometer.Positive substance P(SP)expression,mRNA and relative protein expression in the trigeminal cervical pulp complex(TCC)in rat mice by immunohistochemistry,RT-qPCR and Western blot analysis.Results Compared with the normal group,the model group has macro signs such as red eyes,irritability,cage fighting,dark tongue and stasis points;regular head positioning and frequent hair management;increased distance in the central area(P<0.05).The relative viscosity of whole blood,plasma viscosity,red blood cell aggregation index all increased(P<0.05),and the red blood cell variant index decreased(P<0.05);SP positive expression,mRNA and the relative protein expression increased in TCC(P<0.05).Compared with model group,Zhengtian pill group can improve macro representation and behavior significantly;can significantly reduce central distance(P<0.05);can reduce plasma viscosity,high-shear relative viscosity of whole blood,and red blood cell aggregation index(P<0.05);increase the Red blood cell variant index significantly(P<0.05);down-regulate SP positive expression,mRNA and relative protein expression in TCC(P<0.05,P<0.01).Conclusion SD rats were used for 4 weeks to construct a migraine model of hyperhepatic hyperactivity and stasis,which was basically consistent with the clinical manifestations.

Objective To explore the distribution of common syndromes in sepsis based on literature data mining.Methods The literature related to sepsis symptoms analysis included in China National Knowledge Infrastructure(CNKI),Chinese Biomedical Literature Database(CBM),VIP database(VIP)and Wanfang Data were searched by computer to extract relevant information and establish a database.Analysis of common syndromes and symptoms of sepsis by frequency statistics.Based on LTM-EAST algorithm of two-step hidden tree analysis,a symptom hidden structure model with frequency≥10 was constructed by Latern 5.0 software,and potential syndromes were inferred through comprehensive clustering.Symptoms with frequency＞5%were selected for factor analysis,common factors were extracted for further systematic cluster analysis,and potential syndromes were inferred according to the results.Based on the above results,the common syndromes and clinical characteristics of sepsis were analyzed.Results A total of 792 literatures and 961 syndromes records and 877 symptoms records were included,involving 48 syndromes after the regulation,with high frequency of pattern of blood stasis,pattern of yangming fu-organ excess,pattern of internal block of the heart orifices due to phlegm and heat,pattern of yang prostration,etc.The frequencies were 12.70%(122/961),9.68%(93/961),9.47%(91/961)and 7.80%(75/961).It involved 179 symptoms after the regulation,including high fever,red tongue,thirst,dysphoria,etc.The frequencies were 27.82%(244/877),27.36%(240/877),27.25%(239/877),25.54%(244/877).Forty-three hidden variables were obtained through implicit structure analysis,according to professional knowledge,12 potential syndromes were obtained by comprehensive clustering,including pattern of blood stasis,pattern of yangming fu-organ excess,pattern of phlegm and heat congesting in the lung,pattern of fire toxin entering the nutrient phase,pattern of Yin deficiency with exuberant heat,pattern of Yang prostration,pattern of Qi deficiency,pattern of Yang deficiency,pattern of turbid phlegm obstructing the lung,pattern of damp-heat encumbering the spleen,pattern of intense exuberant heat toxin,pattern of pathogenic factors repressing defensive Qi of lung.Twenty-five common factors were obtained by factor analysis,and 12 potential syndromes were inferred by systematic clustering,including pattern of heart-Yang deficiency,pattern of phlegm and heat congesting in the lung,pattern of yangming fu-organ excess,pattern of Yang prostration,pattern of intense exuberant heat toxin,pattern of Yang deficiency in spleen and kidney,pattern of Qi deficiency,pattern of phlegm and heat damaging Yin,pattern of Qi deficiency with blood stasis,pattern of Qi and Yin deficiency,pattern of pathogenic factors repressing defensive Qi of lung,and pattern of heat sinking into the pericardium.According to the standard of common traditional Chinese medicine(TCM)syndromes and their symptom characteristics,9 common syndromes of sepsis were finally obtained,including pattern of intense exuberant heat toxin,blood stasis,pattern of yangming fu-organ excess,pattern of phlegm and heat congesting in the lung,pattern of fire toxin entering,pattern of pathogenic factors repressing defensive Qi of lung,pattern of Qi and Yin deficiency,pattern of Qi deficiency,pattern of Yang prostration.Conclusion The symptoms and characteristics of common syndromes of sepsis are significant,which can provide evidence for the syndrome diagnosis standard of sepsis.

Objective To explore the distribution of traditional Chinese medicine basic syndromes in sepsis,and to provide evidence for the establishment of diagnostic criteria of sepsis syndromes.Methods Literatures related to sepsis syndrome included in China National Knowledge Infrastructure(CNKI),Wanfang Data,VIP database(VIP)and Chinese Biomedical Literature Database(CBM)databases were searched from the establishment of the database to May 30,2020.Screen the literature and extract data,establish a database for statistical description and analysis of syndrome elements,basic syndromes and symptoms.Analyze the association rule of syndrome elements based on the Apriori algorithm.Based on the two-step hidden tree analysis LTM-EAST algorithm,a symptom latent structure model was constructed,and a comprehensive cluster analysis and model interpretation were performed.Results A total of 383 articles related to sepsis syndromes were included,involving 31 basic syndromes,146 symptoms and 29 syndromes elements.The basic syndromes with frequencies≥5%and cumulative composition ratios≥75%were heat and toxin syndrome,blood stasis syndrome,Yin deficiency syndrome,heat and closing syndrome,Qi deficiency syndrome,phlegm and heat syndrome,Yang prostration syndrome,Fu-organ excess syndrome,and Yang deficiency syndrome.Perform the association rule analysis on syndrome elements with a frequency＞5,obtaining 8 strong association rules,and inferring 7 basic syndromes,including Fu-organ excess syndrome,heat and toxin syndrome,heat and closing syndrome,phlegm clouding the heart syndrome,heat disturbing the heart spirit syndrome,phlegm and heat syndrome,phlegm block syndrome.The symptoms with frequency＞5 were analyzed by hidden structure,41 hidden variables and 82 hidden categories were obtained,and 11 comprehensive clustering models were obtained through comprehensive clustering.Eleven basic syndromes were inferred,including heat and toxin syndrome,blood stasis syndrome,Yin deficiency syndrome,heat and closing syndrome,Qi deficiency syndrome,phlegm and heat syndrome,Yang prostration syndrome,Fu-organ excess syndrome,Yang deficiency syndrome,Yingfen syndrome,and phlegm-dampness syndrome.Combined with all of methods above,9 basic syndromes,heat and toxin syndrome,blood stasis syndrome,Yin deficiency syndrome,heat and closing syndrome,Qi deficiency syndrome,phlegm and heat syndrome,Yang prostration syndrome,Fu-organ excess syndrome,and Yang deficiency syndrome were finally confirmed.Conclusion There are 9 common basic syndromes of sepsis,and the sufficient syndromes are mainly heat and toxin syndrome,blood stasis syndrome,heat and closing syndrome,phlegm and heat syndrome and Fu-organ excess syndrome,while the deficiency syndromes are mainly Yin deficiency syndrome,Qi deficiency syndrome,Yang prostration syndrome,and Yang deficiency syndrome,with each basic syndrome having certain symptom characteristics.

Objective:To investigate the differential expression of four-jointed box kinase 1 (FJX1) gene in colorectal cancer and its relationship with prognosis and the related mechanisms.Methods:On July 16, 2021, the transcriptome data and clinical data of colorectal cancer were downloaded from The Cancer Genome Atlas (TCGA) database to analyze the expressions of FJX1 mRNA in colorectal cancer tissues and paracancerous tissues, and the relationship between FJX1 mRNA and clinicopathological characteristics and prognosis of patients. Receiver operating characteristic (ROC) curve was drawn to evaluate the value of FJX1 mRNA in predicting the survival of patients with colorectal cancer. Cox proportional hazards model was used to evaluate whether FJX1 mRNA was an independent influencing factor for prognosis of colorectal cancer. The overall survival (OS) time and survival status of colorectal cancer patients were downloaded from the Gene Expression Omnibus (GEO) database, and the relationship between FJX1 mRNA and prognosis of patients was analyzed. The methylation data of colorectal cancer was downloaded from the University of California, Santa Cruz (UCSC xena) database to determine the degree of methylation at each site of FJX1 mRNA and the correlation between the expression of FJX1 mRNA and the degree of methylation at each site. Signaling pathways associated with FJX1 mRNA in colorectal cancer were analyzed by using the Gene Set Enrichment Analysis (GSEA) (4.1.0). The correlation between FJX1 mRNA and tumor-infiltrating immune cells was investigated by using the Tumor Immunity Evaluation Resource (TIMER) database. Spearman analysis and small molecule/drug sensitivity analysis were used to explore the correlation between FJX1 mRNA expression and drug sensitivity.Results:In the transcriptome data of 612 colorectal cancer cases in TCGA database, the expression of FJX1 mRNA in colorectal cancer tissues was higher than that in the paracancerous tissues ( P < 0.001). In 549 colorectal cancer patients with complete data, FJX1 mRNA expression was correlated with M stage ( P = 0.007), pathological stage (stage Ⅳ vs. stage Ⅰ, P = 0.016; stage Ⅳ vs. stage Ⅱ, P = 0.03; stage Ⅳ vs. stage Ⅲ, P = 0.012), but it was not correlated with age, gender, T stage and N stage (all P > 0.05). In TCGA database and GEO database, the patients were divided into high expression group and low expression group according to the median expression of FJX1 mRNA. The OS in FJX1 mRNA high expression group was worse than that in low expression group (all P<0.05). The ROC curve of FJX1 mRNA expression on the 1-, 3-, and 5-year OS rates of colorectal cancer patients was drawn by using the data in TCGA database, and the areas under the curve (AUC) were 0.595, 0.625 and 0.764, respectively. Multivariate Cox regression analysis showed that age ( HR = 1.050, 95% CI 1.028-1.073, P < 0.001), T stage ( HR = 1.787, 95% CI 1.090-2.927, P = 0.021) and high FJX1 mRNA expression ( HR = 1.160, 95% CI 1.049-1.282, P = 0.004) were independent influencing factors for poor OS in colorectal cancer. The gene set enrichment analysis found that FJX1 mRNA was related to colorectal cancer, TGF-β signaling pathway, VEGF signaling pathway, Wnt signaling pathway, etc. The expression of FJX1 mRNA in colon cancer was negatively correlated with the degree of methylation of FJX1 mRNA ( r = -0.16, P < 0.001), and the expression of FJX1 mRNA in rectal cancer was positively correlated with the degree of methylation of FJX1 mRNA ( r = 0.33, P < 0.001). The expression of FJX1 mRNA was related to the infiltration of resting memory CD4 + T cells, M0 macrophages and resting dendritic cells. FJX1 mRNA was significantly associated with the resistance of various chemotherapeutic drugs and tumor-targeted drugs such as methotrexate, 5-fluorouracil, gefitinib, etc. Conclusions:FJX1 mRNA may be a potential biomarker of colorectal cancer and is associated with the infiltration of immune cells.

OBJECTIVE To expl ore the mechanism of effe ctive fractions from Xiongma decoction in the treatment of migraine with hyperactivity of liver-yang and blood stasis. METHODS Totally 70 male SD rats were randomly divided into normal group ， model group ，positive control group （Flunarizine hydrochloride capsules 0.9 mg/kg），low-dose and high-dose groups of Xiongma decoction effective fractions （ethyl acetate extract 0.87，3.46 g/kg，n-butanol extract 1.80，7.20 g/kg）. Except for normal group ， rats in other groups were given aconite decoction （2 g/kg），once a day ，for 4 consecutive weeks to establish the hyperactivity model of liver-yang. On the 15th day of modeling ，all administration groups were given corresponding drugs intragastrically at the same time ，once a day ，for 2 consecutive weeks. On the 29th day of modeling ，rats trigeminal ganglion was stimulated to establish the migraine model with hyperactivity of liver-yang and blood stasis ，then the medication was maintained for another time according to the above method. The macroscopic signs and behavior of the rats were observed ；positive expression ，mRNA and protein expression of transient receptor potential vanilloid 1 （TRPV1），calcitonin generelated peptide （CGRP），calcitonin receptor-like receptor （CRLR） and receptor-associated membrane proteins 1 （RAMP1） in trigeminal cervical spinal complex （TCC） were detected by immunohistochemistry ，RT-qPCR and Western blot assay. RESULTS Rats in model group showed macrophysical signs and behaviora l manifestations related to migraine with hyperactivity of liver-yang and blood stasis. Thirty minutes after last administration ，the above conditions of rats in Xiongma decoction effective fraction groups were improved significantly. Compared with normal group ， positive expression，mRNA and protein expression of TRPV 1，CGRP， CRLR and RAMP 1 in TCC of rats in the model group were significantly increased （P＜0.05）. Compared with model zengguirong@hnse.org group， most of above indicators in Xiongma decoction effective fraction groups were significantly decreased （P＜0.05）. CONCLUSIONS The mechanism of Xiongma decoction in preventing and treating migraine with hyperactivity of liver-yang and blood stasis may be related to inhibit the activity of TRPV1-CGRP/CGRP receptor signaling pathway in TCC.