Objective In this study,a risk prediction model for Alzheimer's disease(AD)was constructed based on Alzheimer's disease neuroimaging initiative(ADNI)database to provide a theoretical basis for a better understanding of AD,as well as to provide a reference basis for the rational allocation of health resources and the development of preventive and control strategies.Methods The ADNIMERGE,UPENNBIOMK,ADNINIGHTINGALELONG and DXSUM datasets from the ADNI database were downloaded and random forest algorithm was used for the missing values imputation.Variable screening was performed by LASSO regression.Borderline-1 SMOTE resampling was used to adjusted for intergroup balance,and the variables were incorporated into base classifiers including XGBoost,random forest,bagging,AdaBoost,and support vector machine.Enhanced integrated classifiers were then constructed based on voting and Stacking strategies.Youden index,area under curve(AUC),F-score,G-mean,accuracy,Matthews correlation coefficient(MCC)and Kappa on the validation set were used to evaluate and compare the model efficacy.Results The performance of the classifiers based on balanced data were improved for both base and enhanced integrated classifiers.The performance of stacking and voting enhanced integrated classifiers constructed based on advantageous base classifiers have better performance compared to the base classifiers.After data balancing,the XGBoost performed better in the base classifiers(AUC:0.9090,accuracy:0.9091)and voting algorithm performs better in enhanced integrated classifiers(AUC:0.9178,accuracy:0.9179).Conclusion After Borderline-1 SMOTE resampling adjusted,the performance of both base classifiers and the enhanced integrated classifiers were all improved.For balanced data,XGBoost classifier and the voting enhanced integrated classifier can effectively assist in clinical prediction of Alzheimer's disease progression.

Objective:To explore the causal association between insulin-like growth factor-1(IGF-1)and colorectal cancer(CRC)based on two sample Mendelian randomization(MR)analysis.Methods:A bidirectional two sample MR analysis was conducted based on publicly aggregated data from the IEU OpenGWAS project.The inverse variance weighted(IVW)method was used as the main analysis model to assess the causal relationship between IGF-1 and CRC.Additional analyses were performed using weighted median(WM),MR-Egger regression,weighted mode estimator(WME),and simple mode(SM)methods.Sensitivity analysis was performed to assess the robustness of the results.Results:A total of 386 single nucleotide polymorphisms(SNPs)were selected as instrumental variables(IVs)with IGF-1 as the exposure factor.The MR analysis results revealed a positive causal association between IGF-1 and the risk of CRC[odds ratio(OR)=1.178,95％confidence interval(CI):1.092-1.272)](P＜0.001),and the association remained significant after adjusting for height[OR(95％CI)=1.214(1.111,1.327)](P＜0.001).Cochran's Q-test showed heterogeneity among the IVs(P＜0.05),while the horizontal pleiotropy of IV was not detected by the MR-Egger regression(P＞0.05).The leave-one-out analysis showed that the MR results were robust.Reverse MR analysis indicated no reverse causal relationship between IGF-1 and CRC[OR(95％CI):1.017(0.997,1.037)](P=0.103).Conclusion:There is a causal relationship between IGF-1 level and CRC,and elevated IGF-1 level could be a risk factor for CRC.

Objective:To explore the correlation between skeletal muscle mass and strength with metabolic syndrome in children.Methods:This cross-sectional study was conducted involving 383 children aged 10 to 15 years who visited the Department of Child Health Care, Children′s Hospital of Nanjing Medical University from June 2021 to December 2022. Their height, weight, waist circumference, body composition, grip strength and blood pressure were measured. Relative skeletal muscle mass, muscle-to-fat ratio, and grip strength-to-body weight index were calculated. The levels of fasting blood glucose, lipids and insulin were tested. Homeostasis model assessment of insulin resistance (HOMA-IR) was calculated. Children were divided into the metabolic syndrome group and the non-metabolic syndrome group. Independent t test or Mann-Whitney U test etc. was used to compare the difference between groups. Spearman correlation analysis and binary Logistic regression were used to investigate the correlation between skeletal muscle mass and strength and metabolic syndrome. The area under the curve (AUC) of receiver operating characteristic (ROC) curve was used to compare the accuracy of the index of skeletal muscle in predicting metabolic syndrome in children. Results:Among the 383 children, 282 (73.6%) were male, at the age of 11.4 (10.6, 12.5) years. There were 216 children (56.4%) diagnosed with obesity and 90 children (23.5%) diagnosed with metabolic syndrome. Relative skeletal muscle mass, muscle-to-fat ratio, and grip strength-to-body weight index of the metabolic syndrome group were all lower than those in the non-metabolic syndrome group (all P<0.001). After adjusting for sex and age, relative skeletal muscle mass, muscle-to-fat ratio, and grip strength-to-body weight index were all negatively correlated with body mass index ( r=-0.84, -0.38, -0.63), waist circumference ( r=-0.76, -0.36, -0.70), systolic blood pressure ( r=-0.42, -0.21, -0.38), diastolic blood pressure ( r=-0.33, -0.18, -0.24), triglycerides ( r=-0.29, -0.13, -0.23), fasting insulin ( r=-0.28, -0.20, -0.29), and HOMA-IR ( r=-0.26, -0.18, -0.26) (all P<0.05), and positively correlated with high-density lipoprotein cholesterol ( r=0.38, 0.13, 0.31, all P<0.01). After adjusting for sex and age, high relative skeletal muscle mass, high muscle-to-fat ratio, and high grip strength-to-body weight index all decreased the risks of metabolic syndrome ( OR=0.80, 0.55, 0.90), obesity ( OR=0.53, 0.64, 0.82), hypertension ( OR=0.86, 0.58, 0.92), low high-density lipoprotein cholesterol ( OR=0.83, 0.62, 0.92), hypertriglyceridemia ( OR=0.88, 0.78, 0.96). After adjusting for sex and age, high relative skeletal muscle mass and high grip strength-to-body weight index all decreased the risks of hyperglycemia ( OR=0.93 and 0.95, all P<0.05). ROC curve analysis showed that the relative skeletal muscle mass, muscle-to-fat ratio, and grip strength-to-body weight index all had good predictive accuracy of metabolic syndrome in children (AUC=0.79, 0.71, 0.76), with optimal cutoff values of 40%, 1.2, and 35%, respectively. Conclusions:High relative skeletal muscle mass, high muscle-to-fat ratio, and high grip strength-to-body weight index are all protective factors for metabolic syndrome in children. Regular measurement of skeletal muscle mass and grip strength can aid in the early identification and prevention of obesity and metabolic syndrome during childhood .

Objective:This study aims to review the practices and approaches of the European Network of Research Integrity Offices(ENRIO) in strengthening the construction and governance of research integrity, providing references for China.Methods:Through literature review and official website information, summarized the basic situation and practical experience of ENRIO.Results:ENRIO had carried out a wide range of practical activities in research integrity, including establishing a network system, forming working groups, developing guidelines, providing training and education, and creating an international communication platform.Conclusions:ENRIO has greatly benefited the construction of research integrity in Europe. These measures have also provided references for China. China′s research integrity construction is a long and arduous task. It is recommended to combine China′s actual situation, empower professional bodies, establish a mechanism for whistleblower protection and investigation and handling, strengthen education and training, and enhance research and international cooperation.

OBJECTIVE: To observe the effects of Tiaoshu Anshen (regulating the hinge and calming the mind) acupuncture on sleep quality and serum levels of 5-hydroxytryptamine (5-HT) and dopamine (DA) in patients with chronic insomnia. METHODS: A total of 58 patients with chronic insomnia were randomly divided into an acupuncture group and a medication group, 29 cases in each group. Tiaoshu Anshen acupuncture was applied at Baihui (GV20) and bilateral Shenmen (HT7), Sanyinjiao (SP6), Benshen (GB13) in the acupuncture group, once a day, 1-day interval was taken after 6 consecutive days of treatment. Estazolam tablet was given orally before bed in the medication group, 1 mg each time. The 4-week treatment was required in both groups. Before and after treatment, the sleep quality was assessed by Pittsburgh sleep quality index (PSQI) and polysomnography (PSG), the serum levels of 5-HT and DA were detected by ELISA. RESULTS: After treatment, the item scores and total scores of PSQI were decreased compared with those before treatment in the two groups (P<0.05); in the acupuncture group, the scores of sleep quality, sleep latency, sleep time, sleep efficiency, sleep disorders and total score of PSQI were lower than those in the medication group (P<0.05). After treatment, the total sleep time (TST) was prolonged (P<0.05), the sleep latency (SL) and wake after sleep onset (WASO) were shortened (P<0.05), the sleep efficiency (SE%), percentage of non-rapid eye movement stage 3 (N3%), percentage of rapid eye movement stage (REM%) and serum levels of 5-HT were increased (P<0.05) compared with those before treatment; the percentage of non-rapid eye movement stage 1 (N1%), percentage of non-rapid eye movement stage 2 (N2%) and serum levels of DA were decreased (P<0.05) compared with those before treatment in the two groups. After treatment, in the acupuncture group, TST was longer, while SL and WASO were shorter than those in the medication group (P<0.05), SE%, N3%, REM% and serum level of 5-HT were higher, while N1%, N2% and serum level of DA were lower than those in the medication group (P<0.05). CONCLUSION Tiaoshu Anshen acupuncture may improve the sleep quality by regulating the serum neurotransmitter levels i.e. 5-HT and DA in patients with chronic insomnia.
Humans ; Sleep Initiation and Maintenance Disorders/physiopathology* ; Male ; Acupuncture Therapy ; Female ; Middle Aged ; Adult ; Serotonin/blood* ; Sleep Quality ; Acupuncture Points ; Dopamine/blood* ; Aged ; Neurotransmitter Agents/blood* ; Young Adult

Objective:To explore the mechanisms of action of Huanglian Jiedu decoction combined with Xijiao Dihuang decoction (HLJDT-XJDH) in regulating fibroblasts in the treatment of psoriasis. Methods:A mouse model of psoriasis was established by topical application of imiquimod 5% cream on the shaved back; HLJDT-XJDH at different doses of 7.7 and 30.6 g/kg was administered via gavage for intervention, and methotrexate (2 mg/kg) served as a positive control; after 7 days, the severity of skin lesions was assessed using the psoriasis area and severity index (PASI), while histopathological changes of skin tissues were evaluated using hematoxylin-eosin (HE) staining and Baker scoring. For in vitro experiments, fibroblasts were divided into a control group, a model group, a low-dose (5% drug-containing serum) intervention group, and a high-dose (20% drug-containing serum) intervention group; cells in the control group were cultured with 20% normal rat serum for 24 hours; in the model group, cells cultured with 20% normal rat serum were stimulated with 5 ng/ml tumor necrosis factor (TNF) -α and 50 ng/ml interleukin (IL) -17A for 24 hours to mimic fibroblasts during the occurrence of psoriasis; cells in the low- and high-dose intervention groups received the same stimulation as the model group, and were cultured for 24 hours with 5% and 20% HLJDT-XJDH-containing serum, respectively, but not with the 20% normal rat serum. After the above treatment, these cells were co-cultured with keratinocytes (HaCaT cells) using a Transwell system. In addition, on the basis of the control group, fibroblasts were divided into the model group, 20% drug-containing serum intervention group, and 20% drug-containing serum intervention + OE-SFRP2 group; TNF-α and IL-17A were used to stimulate the cells to simulate the psoriatic state; the treatment in the 20% drug-containing serum intervention group was carried out as previously described; in the 20% drug-containing serum intervention + OE-SFRP2 group, cells were transfected with the vector for 48 hours to establish an overexpression model, followed by culture with 20% drug-containing serum for 24 hours, without co-culture with HaCaT cells.. Cell counting kit-8 (CCK-8) assay was performed to assess cell viability, flow cytometry to measure apoptosis rates, enzyme-linked immunosorbent assay (ELISA) to detect levels of inflammatory cytokines (TNF-α, IL-1β, IL-6) as well as chemokine ligand (CXCL) 1 and CXCL12 in mouse serum or cell culture supernatant, qPCR to determine the mRNA expression of inflammatory cytokines, chemokines, cell cycle- and proliferation-related factors, as well as SFRP2 in mouse skin tissues or cells, and Western blot analysis to determine the protein expression of SFRP2, Wnt3a, and β-catenin in fibroblasts. One-way analysis of variance was employed for intergroup comparisons, and post-hoc analysis was conducted using Tukey's test. Results:In vivo mouse experiments showed that compared with the normal control group, the model group exhibited typical psoriatic characteristics in skin morphology, including significant inflammatory infiltration in skin tissues and marked epidermal thickening; compared with the normal control group, the serum levels of TNF-α (531.16 ± 28.27 pg/ml vs. 239.58 ± 10.39 pg/ml), IL-1β (111.40 ± 5.16 pg/ml vs. 80.35 ± 3.87 pg/ml), and IL-6 (109.17 ± 4.84 pg/ml vs. 71.73 ± 2.04 pg/ml) significantly increased in the model group, along with their mRNA expression levels in mouse skin tissues (all P < 0.001) ; compared with the model group, the treatment group showed alleviated psoriatic manifestations, and significant reductions in the levels of inflammatory factors TNF-α (low-dose, high-dose, and positive control groups: 420.80 ± 29.30 pg/ml, 322.33 ± 9.40 pg/ml, 322.97 ± 12.16 pg/ml, respectively), IL-1β (98.69 ± 4.49 pg/ml, 89.02 ± 1.56 pg/ml, 88.88 ± 2.08 pg/ml, respectively), and IL-6 (94.07 ± 3.76 pg/ml, 80.54 ± 3.30 pg/ml, 83.21 ± 3.18 pg/ml, respectively), as well as in their mRNA expression levels (all P < 0.001). In in vitro fibroblast experiments, compared with the control group, the model group exhibited a significant elevation in the supernatant levels of IL-1β (126.42 ± 3.56 pg/ml vs. 34.81 ± 0.44 pg/ml), IL-6 (459.44 ± 9.35 pg/ml vs. 115.51 ± 7.26 pg/ml), CXCL1 (2 434.88 ± 127.63 pg/ml vs. 762.85 ± 30.60 pg/ml) and CXCL12 (3 542.14 ± 35.86 pg/ml vs. 2 095.86 ± 45.12 pg/ml), the expression levels of their mRNAs (all P < 0.001), as well as the protein expression levels of SFRP2, Wnt3a, and β-catenin; after intervention with HLJDT-XJDH-containing serum, all the above indices significantly decreased (all P < 0.001). However, when 20% drug-containing serum intervention was administered simultaneously, the expression of inflammatory factors and chemokines in fibroblasts was significantly higher in the SFRP2 overexpression group than in the non-overexpression group (all P < 0.01). When fibroblasts were co-cultured with HaCaT cells, the model group showed significantly increased cell viability but a decreased apoptosis rate of HaCaT cells compared with the control group, while the low- and high-dose intervention groups showed significantly decreased cell viability but increased apoptosis rates of HaCaT cells compared with the model group (all P < 0.05) . Conclusion:HLJDT-XJDH may exert therapeutic effects in psoriasis by downregulating the SFRP2/Wnt/β-catenin signaling pathway, thereby inhibiting fibroblast activation and inflammatory process, which subsequently suppresses the proliferation of keratinocytes and the activation of inflammatory cells.

Objective The purpose of this study is to screen the key genes and clinical characteristics related to the death or prognosis of lung adenocarcinoma(LUAD)patients based on the cancer genome atlas(TCGA)database,then construct and verify the effect of LUAD risk prediction model.Methods Clinical information and RNA sequencing data of lung adenocarcinoma patients were extracted from TCGA database.The deferentially expressed genes were screened,and hub genes were selected by protein interaction(PPI)network.70%of the data was used as the training set,and the entire data set was used as the validation set.In the training set,elastic net regression analysis was used to select prognostic genes and clinical characteristics,and Cox multivariate regression analysis was used to build a risk prediction model.The predictive performance of the model was evaluated by the area under the receiver's operating characteristic curve(AUC),C－index,and calibration curve.And the effect of the model was verified in the validation set.Results Elastic net regression analysis identified 23 factors associated with the survival status of LUAD patients.The variables that finally included in the predictive model include 6 genes(SEC61A1(P＝0.004),MAP2K1(P＝0.026),MMP1(P＝0.001),SLC2A1(P＝0.010),B4GALT1(P<0.001),ERO1A(P＝0.024)),and M stage(P＝0.003),N stage(P<0.001).In training set and test set,AUC was 0.764 and 0.710,C－ index was 0.732 and 0.704,respectively.The tdROC curve and calibration curve showed that the predicted values of the model were highly consistent with the actual observed values.The Kaplan－Meier survival curve showed that the survival time of patients in low－risk group was statistically significantly longer than that of those in high－risk group(P<0.05).Conclusion Low expression of 2 genes(SEC61A1,MAP2K1),high expression of 4 genes(MMP1,SLC2A1,B4GALT1 and ERO1A),and distant metastasis of the primary tumors and the deepening of lymph node metastasis resulted in a significantly shorter survival time in LUAD patients.The prognosis analysis model based on elastic network has satisfactory predictive ability,which can provide scientific basis for prediction of the death risk of LUAD.

Objective:To explore the mechanisms of action of Huanglian Jiedu decoction combined with Xijiao Dihuang decoction (HLJDT-XJDH) in regulating fibroblasts in the treatment of psoriasis. Methods:A mouse model of psoriasis was established by topical application of imiquimod 5% cream on the shaved back; HLJDT-XJDH at different doses of 7.7 and 30.6 g/kg was administered via gavage for intervention, and methotrexate (2 mg/kg) served as a positive control; after 7 days, the severity of skin lesions was assessed using the psoriasis area and severity index (PASI), while histopathological changes of skin tissues were evaluated using hematoxylin-eosin (HE) staining and Baker scoring. For in vitro experiments, fibroblasts were divided into a control group, a model group, a low-dose (5% drug-containing serum) intervention group, and a high-dose (20% drug-containing serum) intervention group; cells in the control group were cultured with 20% normal rat serum for 24 hours; in the model group, cells cultured with 20% normal rat serum were stimulated with 5 ng/ml tumor necrosis factor (TNF) -α and 50 ng/ml interleukin (IL) -17A for 24 hours to mimic fibroblasts during the occurrence of psoriasis; cells in the low- and high-dose intervention groups received the same stimulation as the model group, and were cultured for 24 hours with 5% and 20% HLJDT-XJDH-containing serum, respectively, but not with the 20% normal rat serum. After the above treatment, these cells were co-cultured with keratinocytes (HaCaT cells) using a Transwell system. In addition, on the basis of the control group, fibroblasts were divided into the model group, 20% drug-containing serum intervention group, and 20% drug-containing serum intervention + OE-SFRP2 group; TNF-α and IL-17A were used to stimulate the cells to simulate the psoriatic state; the treatment in the 20% drug-containing serum intervention group was carried out as previously described; in the 20% drug-containing serum intervention + OE-SFRP2 group, cells were transfected with the vector for 48 hours to establish an overexpression model, followed by culture with 20% drug-containing serum for 24 hours, without co-culture with HaCaT cells.. Cell counting kit-8 (CCK-8) assay was performed to assess cell viability, flow cytometry to measure apoptosis rates, enzyme-linked immunosorbent assay (ELISA) to detect levels of inflammatory cytokines (TNF-α, IL-1β, IL-6) as well as chemokine ligand (CXCL) 1 and CXCL12 in mouse serum or cell culture supernatant, qPCR to determine the mRNA expression of inflammatory cytokines, chemokines, cell cycle- and proliferation-related factors, as well as SFRP2 in mouse skin tissues or cells, and Western blot analysis to determine the protein expression of SFRP2, Wnt3a, and β-catenin in fibroblasts. One-way analysis of variance was employed for intergroup comparisons, and post-hoc analysis was conducted using Tukey's test. Results:In vivo mouse experiments showed that compared with the normal control group, the model group exhibited typical psoriatic characteristics in skin morphology, including significant inflammatory infiltration in skin tissues and marked epidermal thickening; compared with the normal control group, the serum levels of TNF-α (531.16 ± 28.27 pg/ml vs. 239.58 ± 10.39 pg/ml), IL-1β (111.40 ± 5.16 pg/ml vs. 80.35 ± 3.87 pg/ml), and IL-6 (109.17 ± 4.84 pg/ml vs. 71.73 ± 2.04 pg/ml) significantly increased in the model group, along with their mRNA expression levels in mouse skin tissues (all P < 0.001) ; compared with the model group, the treatment group showed alleviated psoriatic manifestations, and significant reductions in the levels of inflammatory factors TNF-α (low-dose, high-dose, and positive control groups: 420.80 ± 29.30 pg/ml, 322.33 ± 9.40 pg/ml, 322.97 ± 12.16 pg/ml, respectively), IL-1β (98.69 ± 4.49 pg/ml, 89.02 ± 1.56 pg/ml, 88.88 ± 2.08 pg/ml, respectively), and IL-6 (94.07 ± 3.76 pg/ml, 80.54 ± 3.30 pg/ml, 83.21 ± 3.18 pg/ml, respectively), as well as in their mRNA expression levels (all P < 0.001). In in vitro fibroblast experiments, compared with the control group, the model group exhibited a significant elevation in the supernatant levels of IL-1β (126.42 ± 3.56 pg/ml vs. 34.81 ± 0.44 pg/ml), IL-6 (459.44 ± 9.35 pg/ml vs. 115.51 ± 7.26 pg/ml), CXCL1 (2 434.88 ± 127.63 pg/ml vs. 762.85 ± 30.60 pg/ml) and CXCL12 (3 542.14 ± 35.86 pg/ml vs. 2 095.86 ± 45.12 pg/ml), the expression levels of their mRNAs (all P < 0.001), as well as the protein expression levels of SFRP2, Wnt3a, and β-catenin; after intervention with HLJDT-XJDH-containing serum, all the above indices significantly decreased (all P < 0.001). However, when 20% drug-containing serum intervention was administered simultaneously, the expression of inflammatory factors and chemokines in fibroblasts was significantly higher in the SFRP2 overexpression group than in the non-overexpression group (all P < 0.01). When fibroblasts were co-cultured with HaCaT cells, the model group showed significantly increased cell viability but a decreased apoptosis rate of HaCaT cells compared with the control group, while the low- and high-dose intervention groups showed significantly decreased cell viability but increased apoptosis rates of HaCaT cells compared with the model group (all P < 0.05) . Conclusion:HLJDT-XJDH may exert therapeutic effects in psoriasis by downregulating the SFRP2/Wnt/β-catenin signaling pathway, thereby inhibiting fibroblast activation and inflammatory process, which subsequently suppresses the proliferation of keratinocytes and the activation of inflammatory cells.

Objective The purpose of this study is to screen the key genes and clinical characteristics related to the death or prognosis of lung adenocarcinoma(LUAD)patients based on the cancer genome atlas(TCGA)database,then construct and verify the effect of LUAD risk prediction model.Methods Clinical information and RNA sequencing data of lung adenocarcinoma patients were extracted from TCGA database.The deferentially expressed genes were screened,and hub genes were selected by protein interaction(PPI)network.70%of the data was used as the training set,and the entire data set was used as the validation set.In the training set,elastic net regression analysis was used to select prognostic genes and clinical characteristics,and Cox multivariate regression analysis was used to build a risk prediction model.The predictive performance of the model was evaluated by the area under the receiver's operating characteristic curve(AUC),C－index,and calibration curve.And the effect of the model was verified in the validation set.Results Elastic net regression analysis identified 23 factors associated with the survival status of LUAD patients.The variables that finally included in the predictive model include 6 genes(SEC61A1(P＝0.004),MAP2K1(P＝0.026),MMP1(P＝0.001),SLC2A1(P＝0.010),B4GALT1(P<0.001),ERO1A(P＝0.024)),and M stage(P＝0.003),N stage(P<0.001).In training set and test set,AUC was 0.764 and 0.710,C－ index was 0.732 and 0.704,respectively.The tdROC curve and calibration curve showed that the predicted values of the model were highly consistent with the actual observed values.The Kaplan－Meier survival curve showed that the survival time of patients in low－risk group was statistically significantly longer than that of those in high－risk group(P<0.05).Conclusion Low expression of 2 genes(SEC61A1,MAP2K1),high expression of 4 genes(MMP1,SLC2A1,B4GALT1 and ERO1A),and distant metastasis of the primary tumors and the deepening of lymph node metastasis resulted in a significantly shorter survival time in LUAD patients.The prognosis analysis model based on elastic network has satisfactory predictive ability,which can provide scientific basis for prediction of the death risk of LUAD.

Objective In this study,a risk prediction model for Alzheimer's disease(AD)was constructed based on Alzheimer's disease neuroimaging initiative(ADNI)database to provide a theoretical basis for a better understanding of AD,as well as to provide a reference basis for the rational allocation of health resources and the development of preventive and control strategies.Methods The ADNIMERGE,UPENNBIOMK,ADNINIGHTINGALELONG and DXSUM datasets from the ADNI database were downloaded and random forest algorithm was used for the missing values imputation.Variable screening was performed by LASSO regression.Borderline-1 SMOTE resampling was used to adjusted for intergroup balance,and the variables were incorporated into base classifiers including XGBoost,random forest,bagging,AdaBoost,and support vector machine.Enhanced integrated classifiers were then constructed based on voting and Stacking strategies.Youden index,area under curve(AUC),F-score,G-mean,accuracy,Matthews correlation coefficient(MCC)and Kappa on the validation set were used to evaluate and compare the model efficacy.Results The performance of the classifiers based on balanced data were improved for both base and enhanced integrated classifiers.The performance of stacking and voting enhanced integrated classifiers constructed based on advantageous base classifiers have better performance compared to the base classifiers.After data balancing,the XGBoost performed better in the base classifiers(AUC:0.9090,accuracy:0.9091)and voting algorithm performs better in enhanced integrated classifiers(AUC:0.9178,accuracy:0.9179).Conclusion After Borderline-1 SMOTE resampling adjusted,the performance of both base classifiers and the enhanced integrated classifiers were all improved.For balanced data,XGBoost classifier and the voting enhanced integrated classifier can effectively assist in clinical prediction of Alzheimer's disease progression.