• Communications for Statistical Applications and Methods
• /
• v.17 no.2
• /
• pp.275-292
• /
• 2010

Receiver operating characteristic(ROC) curves can be used to assess the accuracy of tests measured on ordinal or continuous scales. The most commonly used measure for the overall diagnostic accuracy of diagnostic tests is the area under the ROC curve(AUC). When two ROC curves are constructed based on two tests performed on the same individuals, statistical analysis on differences between AUCs must take into account the correlated nature of the data. This article focuses on confidence interval estimation of the difference between paired AUCs. We compare nonparametric, maximum likelihood, bootstrap and generalized pivotal quantity methods, and conduct a monte carlo simulation to investigate the probability coverage and expected length of the four methods.

• Journal of Korean Society for Quality Management
• /
• v.22 no.4
• /
• pp.102-110
• /
• 1994

We investigate the small-sample behavior of the probability of rejection curves and its performance for a equivalence testing procedure based on confidence intervals which was developed with a motivation from bioequivalence studies. This type of equivalence studies are conducted frequently in pharmaceutical industries to compare the relative bioavailabilty of two formulations of a drug and can be applied various fields where assurance of quality equivalence is needed. From the Monte-Carlo simulation results we suggest proper sample sizes for the equivalence testing procedure.

• International Journal of Advanced Culture Technology
• /
• v.8 no.4
• /
• pp.64-76
• /
• 2020

Purpose: Several screening tools have been developed to identify sarcopenia in rural community-dwelling older adults. We aimed to compare the diagnostic accuracy of two such tools, namely the SARC-F and SARC-CalF assessments. Methods: This cross-sectional study on 388 community-dwelling older adults comprised 254 women and 134 men with a mean age of 77.8 ± 6.26 year in Korea. We assessed muscle mass, muscle strength, and physical performance using a bioimpedance analysis device, hydraulic hand dynamometer, and 4 m gait speed test, respectively. Three widely-used diagnostic criteria [the Asian Working Group for Sarcopenia (AWGS), European Working Group on Sarcopenia in Older People, and the International Working Group on Sarcopenia] were applied. Sensitivity and specificity analyses were performed on the SARC-CalF and SARC-F tests. We used receiver-operating characteristic curves and the area under the curves (AUCs) to compare the diagnostic accuracy of the assessments with regard to sarcopenia. Results: An analysis using four sets of diagnostic criteria showed that the prevalence of sarcopenia was 27.6% to 41.0%. Using the AWGS 2019 criteria as a reference standard, the SARC-CalF had a sensitivity of 83.02% and a specificity of 53.71% in the entire study population, whereas the SARC-F had a sensitivity of 79.87% and a specificity of 41.92%. The AUCs for the SARC-CalF and SARC-F tests were 0.725 (95% confidence interval 0.678-0.769) and 0.645 (95% confidence interval 0.595-0.693), respectively (p<001). In the analyses using the other three diagnostic criteria, similarity was also confirmed. Conclusion: SARC-CalF showed better sensitivity than did SARC-F when diagnosing sarcopenia in rural community-dwelling older adults. Further studies are needed to verify this finding in different populations.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.11
• /
• pp.4467-4474
• /
• 2014

Background: We aimed to comprehensively review the evidence for using sputum DNA to detect non-small cell lung cancer (NSCLC). Materials and Methods: We searched PubMed, Science Direct, Web of Science, Chinese Biological Medicine (CBM), Chinese National Knowledge Infrastructure (CNKI), Wanfang, Vip Databases and Google Scholar from 2003 to 2013. The meta-analysis was carried out using a random-effect model with sensitivity, specificity, diagnostic odd ratios (DOR), summary receiver operating characteristic curves (ROC curves), area under the curve (AUC), and 95% confidence intervals (CI) as effect measurements. Results: There were twenty-two studies meeting the inclusion criteria for the meta-analysis. Combined sensitivity and specificity were 0.62 (95%CI: 0.59-0.65) and 0.73 (95%CI: 0.70-0.75), respectively. The DOR was 10.3 (95%CI: 5.88-18.1) and the AUC was 0.78. Conclusions: The overall accuracy of the test was currently not strong enough for the detection of NSCLC for clinical application. Dscovery and evaluation of additional biomarkers with improved sensitivity and specificity from studies rated high quality deserve further attention.

• Imaging Science in Dentistry
• /
• v.41 no.1
• /
• pp.7-10
• /
• 2011

Purpose : The purpose of this study was to compare the conventional and digital radiography in the estimation of working length in mandibular molars. Materials and Methods : Sixty molar teeth were selected and divided into three groups in the basis of canal curves (0-$15^{\circ}$, 15-$30^{\circ}$, > $30^{\circ}$). After the placement of a 15 K-file, radiographs were taken with a conventional film (Fspeed) and a digital sensor. Canal lengths were measured in these images by two observers. Statistical analysis was performed with repeated measures of ANOVA and paired sample t-test with 95% confidence. Results : There was a high inter-observer agreement on the measurements of working length in conventional and digital radiographs. There was no significant difference between the mean values of measurements in conventional and digital radiography. Moreover, there was no significant difference between conventional and digital radiography with the actual values in the basis of canal curves. Conclusion : The accuracy of conventional and digital radiography in the determination of the working length was in an acceptable range.

• Structural Engineering and Mechanics
• /
• v.63 no.2
• /
• pp.259-268
• /
• 2017

In this paper, the incremental nonlinear dynamic analysis is used to evaluate the seismic performance of steel moment frame structures. To this purpose, three special moment frame structure with 5, 10 and 15 stories are designed according to the Iran's national building code for steel structures and the provisions for design of earthquake resistant buildings (2800 code). Incremental Nonlinear Analysis (IDA) is performed for 15 different ground motions, and responses of the structures are evaluated. For the immediate occupancy and the collapse prevention performance levels, the probability that seismic demand exceeds the seismic capacity of the structures is computed based on FEMA350. Also, fragility curves are plotted for three high-code damage levels using HASUS provisions. Based on the obtained results, it is evident that increase in the height of the frame structures reduces the reliability level. In addition, it is concluded that for the design earthquake the probability of exceeding average collapse prevention level is considerably larger than high and full collapse prevention levels.9.

• Imaging Science in Dentistry
• /
• v.33 no.2
• /
• pp.91-95
• /
• 2003

Purpose : To compare the diagnostic accuracy of proximal caries detection between Kodak Insight film and the Biomedisys CDX2000HQ digital (CCD) sensor. Materials and Methods: 156 proximal surfaces of extracted teeth, 78 of which had chemical artificial caries, were used in this study. Four observers interpreted the radiographs using a five-point confidence rating scale to record their diagnoses. The results were analyzed by receiver operating characteristic curves, ANOVA and Kappa values. Result: Analysis using receiver operating characteristic curves revealed the areas under each curve which indicated a diagnostic accuracy of 0.951 in Insight and 0.952 in CDX2000HQ digital sensor. ANOVA revealed no significant differences between the two images with respect to caries detection. Kappa values indicated that the mean intra-observer agreement was 0.85 and inter-observer agreement 0.71 in conventional radiography. In digital radiography, the mean intra-observer agreement was 0.84 and inter-observer agreement 0.72. Conclusion: The results suggest that no significant difference exists between the two modalities for artificial caries detection and that CDX2000HQ was as good as Insight film for this purpose.

• Structural Engineering and Mechanics
• /
• v.67 no.6
• /
• pp.587-601
• /
• 2018

Elevated water tanks are inverted pendulum type structures where drift limit is an important criterion for seismic design and performance evaluation. Explicit drift criteria for elevated water tanks are not available in the literature. In this study, probabilistic approach is used to determine maximum drift limit for damage state of yielding and damage state of collapse for the elevated water tanks supported on RC frame staging. The two damage states are defined using results of incremental dynamic analysis wherein a total of 2160 nonlinear time history analyses are performed using twelve artificial spectrum compatible ground motions. Analytical fragility curves are developed using two-parameter lognormal distribution. The maximum allowable drifts corresponding to yield and collapse level requirements are estimated for different tank capacities. Finally, a single fragility curve is developed which provides maximum drift values for the different probability of damage. Further, for rational consideration of the uncertainties in design, three confidence levels are selected and corresponding drift limits for damage states of yielding and collapse are proposed. These values of maximum drift can be used in performance-based seismic design for a particular damage state depending on the level of confidence.

• The Korean Journal of Physiology and Pharmacology
• /
• v.13 no.5
• /
• pp.367-371
• /
• 2009

The estimation of 90% parametric confidence intervals (CIs) of mean AUC and Cmax ratios in bioequivalence (BE) tests are based upon the assumption that formulation effects in log-transformed data are normally distributed. To compare the parametric CIs with those obtained from nonparametric methods we performed repeated estimation of bootstrap-resampled datasets. The AUC and Cmax values from 3 archived datasets were used. BE tests on 1,000 resampled data sets from each archived dataset were performed using SAS (Enterprise Guide Ver.3). Bootstrap nonparametric 90% CIs of formulation effects were then compared with the parametric 90% CIs of the original datasets. The 90% CIs of formulation effects estimated from the 3 archived datasets were slightly different from nonparametric 90% CIs obtained from BE tests on resampled datasets. Histograms and density curves of formulation effects obtained from resampled datasets were similar to those of normal distribution. However, in 2 of 3 resampled log (AUC) datasets, the estimates of formulation effects did not follow the Gaussian distribution. Bias-corrected and accelerated (BCa) CIs, one of the nonparametric CIs of formulation effects, shifted outside the parametric 90% CIs of the archived datasets in these 2 non-normally distributed resampled log (AUC) datasets. Currently, the 80~125% rule based upon the parametric 90% CIs is widely accepted under the assumption of normally distributed formulation effects in log-transformed data. However, nonparametric CIs may be a better choice when data do not follow this assumption.

• Corrosion Science and Technology
• /
• v.20 no.2
• /
• pp.52-61
• /
• 2021

This study using experimental design and linear regression technique was implemented in order to predict the pitting potential of stainless steel in marine environments, with the target materials being AL-6XN and STS 316L. The various variables (inputs) which affect stainless steel's pitting potential included the pitting resistance equivalent number (PRNE), temperature, pH, Cl^- concentration, sulfate levels, and nitrate levels. Among them, significant factors affecting pitting potential were chosen through an experimental design method (screening design, full factor design, analysis of variance). The potentiodynamic polarization test was performed based on the experimental design, including significant factor levels. From these testing methods, a total 32 polarization curves were obtained, which were used as training data for the linear regression model. As a result of the model's validation, it showed an acceptable prediction performance, which was statistically significant within the 95% confidence level. The linear regression model based on the full factorial design and ANOVA also showed a high confidence level in the prediction of pitting potential. This study confirmed the possibility to predict the pitting potential of stainless steel according to various variables used with experimental linear regression design.