Objective To establish a risk prediction model of diabetic retinopathy (DR) for type 2 diabetic patients (T2DM). Methods A total of 315 T2DM patients (600 eyes) were enrolled in the study. There were 132 males (264 eyes) and 183 females (366 eyes). The mean age was (67.28±12.17) years and the mean diabetes duration was (10.86±7.81) years. The subjects were randomly assigned to model group and check group, each had 252 patients (504 eyes) and 63 patients (126 eyes) respectively. Some basic information including gender, age, education degree and diabetes duration were collected. The probable risk factors of DR including height, weight, blood pressure, fasting glucose, glycosylated hemoglobin (HbA1c), blood urea, serum creatinine, uric acid, triglyceride, total cholesterol, high-density lipoprotein, low density lipoprotein cholesterol and urinary protein. The fundus photograph and the axial length were measured. Multivariate logistic regression was used to analyze the correlative factors of DR and establish the regression equation (risk model). Receiver operating characteristic (ROC) curves were used to determine the cut-off point for the score. The maximum Youden Index was used to determine the threshold of the equation. The check group was used to check the feasibility of the predictive model. Results Among 504 eyes in the model group, 170 eyes were DR and 334 eyes were not. Among 126 eyes in the check group, 45 eyes were DR and 81 eyes were not. Multivariate logistic regression analysis revealed that axial length [β=–0.196, odds ratio (OR)=0.822,P＜0.001], age (β=-0.079,OR=0.924,P＜0.001), diabetes duration (β=0.048,OR=1.049,P=0.001), HbA1c (β=0.184,OR=1.202,P=0.020), urinary protein (β=1.298,OR=3.661,P＜0.001) were correlated with DR significantly and the simplified calculation of the score of DR were as follows:P=7.018–0.196X1–0.079X2+0.048X3+0.148X4+1.298X5 (X1= axial length, X2=age, X3=diabetes duration, X4=glycosylated hemoglobin, X5= urinary protein). The area under the ROC curve for the score DR was 0.800 and the cut-off point of the score was -1.485. The elements of the check group were substituted into the equation to calculate the scores and the scores were compared with the diagnostic threshold to ensure the patients in high-risk of DR. The result of the score showed 84% sensitivity and 59% specificity. ROC curve for the score to predict DR was 0.756. Conclusion Axial length, age, diabetes duration, HbA1c and urinary protein have significant correlation with DR. The sensitivity and specificity of the risk model to predict DR are 84.0% and 59.0% respectively. The area under the ROC curve was 0.756.
The valid results of Meta-analysis on biomedical data, will have an important value to clinical practice and health policy making. To review the validity of meta-analysis results, one should consider the following issues: The coverage ratio of included studies, quality of data, publication bias and its effect, heterogeneity, the correct selection of statistical methods as well as clinical significance and external validity of overall effect size. The results of Meta-analysis will keep on updating as new related studies are located and included.
Objective To assess the completion of the under 5 mortality rate (U5MR) of Millennium Development Goals in 194 member countries of WHO, and to analyze the present situation of the global U5MR. Methods Based on the U5MR and the proportion of main causes of death in the "World Health Statistics 2015", the Millennium Development Goals of the decline of U5MR from 1990 to 2013 was assessed, the U5MR was analyzed by comparison between 2000 and 2013. Bivariate Pearson correlation analysis was used to determine the correlation between mortality and the ratio of infection to non infectious diseases and GDP per person in U5MR. Results By 2013, in 194 WHO member states, the U5MR in 46 (23.71%) countries achieved the millennium development goals. Comparison between 2000 and 2013, there was significant difference between low and high mortality groups in six continents (P<0.05), there was no significant difference between the moderate death groups (P>0.05), there was no significant difference in the ratio of infection to non infectious diseases between the middle and low mortality groups (P>0.05), however there was significant difference between the high mortality groups (P<0.05). There was significant difference in the average decline of U5MR and the ratio of non infectious diseases between low and medium, middle and high mortality groups (P<0.05). The Global U5MR had significant regional differences, the highest U5MR was in Africa, the lowest U5MR was in Europe, the medium U5MR was in North America, Oceania, South America, Asia was becoming the middle level. The U5MR was highly correlated with the ratio of infection to non-infectious diseases in every country (r2000y=0.934,r2013y=0.911,P<0.05), and it was low negatively correlated with GDP per capita (r2000y=–0.443,r2013y=–0.433,P<0.05). Conclusions There is a long way to reduce global child mortality. Prevention and control should focus on Africa and Asia. Prevention and control of infectious diseases is an effective measure for middle and high mortality countries. Prevention and control of non-infectious diseases is an important measure for low mortality countries. Increasing health investment is an important means to further reduce global U5MR.
Through collecting and synthesizing the paper concerning the method of dealing with heterogeneity in the meta analysis, to introduce the multi-levels statistical models, such as meta regression and baseline risk effect model based on random effects, and random effects model based on hierarchical bayes, and to introduce their application of controlling the meta analysis heterogeneity. The multi-levels statistical model will decompose the single random error in the traditional model to data structure hierarchical. Its fitting effect can not only make the meta-analysis result more robust and reasonable, but also guide clinical issues through the interpretation of association variable.
This study aims to determine the salient brain regions with abnormal changes in white matter structures from diffusion tensor imaging (DTI) images of the patients with temporal lobe epilepsy (TLE), and to discriminate the patients with TLE from normal controls (NCs). Firstly, the DTI images from 50 subjects (28 NCs and 22 TLE) were acquired. Secondly, the four measures including the fractional anisotropy (FA), the mean diffusivity (MD), the axial diffusivity (AD) and the radial diffusivity (RD) were calculated. Thirdly, the tract-based spatial statistics (TBSS) was adopted to extract the measures in brain regions with significant differences between the two compared groups. Fourthly, the obtained measures were used as input features of the support vector machine (SVM) for classification, and the support vector machine-recursive feature elimination (SVM-RFE) was compared with the support vector machine-tract-based spatial statistics (SVM-TBSS) method. Finally, the essential brain regions and their spatial distribution were analyzed and discussed. The experimental results showed that the FA measures of the TLE group decreased significantly in the corpus callosum, superior longitudinal fasciculus, corona radiata, external capsule, internal capsule, inferior fronto-occipital fasciculus, fasciculus uncinatus and sagittal stratum, which were nearly bilaterally distributed, while the MD and RD increased significantly in most of these brain regions of the TLE group. Although the AD also increased, the differences were not statistically significant. The SVM-TBSS classifier obtained accuracies of 82%, 76% and 76% using the FA, MD and RD for classification, respectively, and 80% using combined measures. The SVM-RFE classifier obtained accuracies of 90%, 90% and 92% using the FA, MD and RD respectively, while the highest accuracy was 100% using combined measures. These results demonstrated that the SVM-RFE outperformed the SVM-TBSS, and the dominant characteristic influencing classification in brain regions were in associative and commissural fibers. These results illustrated that the measures of DTI images could reveal the abnormal changes in white matter structure of patients with TLE, providing effective information to clarify its pathological mechanism, localize the focus and diagnose automatically.
Objective To investigate confidence interval estimation for the amount of heterogeneity in meta-analysis. Methods On the basis of BT’s method, the approximate Q-statistic distribution following linear transformation of Chi-square was applied to improve the accuracy of Q-statistic distribution, and to obtain the confidence interval for the amount of heterogeneity in meta-analysis. Results In case, the Q1 distribution obtained 95%CI 0.07 to 2.20, while the Q2 distribution obtained 95%CI 0.00 to 1.41; The proposed method Q2 narrowed down the range of confidence interval. Conclusion On account of improving the accuracy of Q-statistic distribution, the proposed method effectively strengthens the coverage probabilities of the confidence interval for the amount of heterogeneity. And the proposed method can also improve the precision of the confidence interval estimation for the amount of heterogeneity.
The WinBUGS software can be called from either R (provided R2WinBUGS as an R package) or Stata software for network meta-analysis. Unlike R, Stata software needs to create relevant ADO scripts at first which simplify operation process greatly. Similar with R, Stata software also needs to load another package when drawing network plots. This article briefly introduces how to implement network meta-analysis using Stata software by calling WinBUGS software.
The selection of summary statistics to use in a meta-analysis is very important for the interpretation and application of its results. This paper introduces some basic concepts of summary statistics in meta-analysis. The selection of a summary statistic for a meta-analysis depends on the following factors: design of the studies being combined, type of data, consistency among the included studies, mathematical properties and ease of interpretation. For continuous data, the weighted mean difference (WMD) is recommended when all trials use the same scale to report their outcomes, while standardized mean difference (SMD) is more appropriate when trials use different scales to report their outcomes, or the means of their outcomes differ greatly. For dichotomous data, rate ratio or relative risk (RR) is bly recommended to be the summary statistics for meta-analyses of randomized trials. The use of odds ratio (OR) as the summary statistic is similar to that of RR, if the event being studied in both the intervention (exposure) and the control group is rare. There is no single measurement that is uniformly best for all meta-analyses.
Statistical analysis of clinical trials has traditionally relied on frequentist methods, but Bayesian statistics has attracted considerable attention from regulators and researchers in recent years due to its unique advantages, and its use in clinical trials is increasing. Despite the obvious advantages of Bayesian statistics, the complexity of its design, implementation and analysis poses a number of challenges to its practical application, which may lead to an increased risk of unregulated use. This study aims to comprehensively sort out the application scenarios, common methods, special considerations and key elements of reporting of Bayesian statistical methods in clinical trials, with the aim of providing researchers with references for conducting Bayesian clinical trials, and promoting the scientific and rational application of Bayesian statistical methods in clinical trials.
The deoxyribonucleic acid (DNA) molecule damage simulations with an atom level geometric model use the traversal algorithm that has the disadvantages of quite time-consuming, slow convergence and high-performance computer requirement. Therefore, this work presents a density-based spatial clustering of applications with noise (DBSCAN) clustering algorithm based on the spatial distributions of energy depositions and hydroxyl radicals (·OH). The algorithm with probability and statistics can quickly get the DNA strand break yields and help to study the variation pattern of the clustered DNA damage. Firstly, we simulated the transportation of protons and secondary particles through the nucleus, as well as the ionization and excitation of water molecules by using Geant4-DNA that is the Monte Carlo simulation toolkit for radiobiology, and got the distributions of energy depositions and hydroxyl radicals. Then we used the damage probability functions to get the spatial distribution dataset of DNA damage points in a simplified geometric model. The DBSCAN clustering algorithm based on damage points density was used to determine the single-strand break (SSB) yield and double-strand break (DSB) yield. Finally, we analyzed the DNA strand break yield variation trend with particle linear energy transfer (LET) and summarized the variation pattern of damage clusters. The simulation results show that the new algorithm has a faster simulation speed than the traversal algorithm and a good precision result. The simulation results have consistency when compared to other experiments and simulations. This work achieves more precise information on clustered DNA damage induced by proton radiation at the molecular level with high speed, so that it provides an essential and powerful research method for the study of radiation biological damage mechanism.