Nonparametric estimation of distributions and diagnostic accuracy based on group-tested results with differential misclassification.
Ontology highlight
ABSTRACT:
SUBMITTER: Zhang W
PROVIDER: S-EPMC8581970 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description: Not available
| S-BSST724 | biostudies-other
Project description:Disorders that share genetic risk factors often are placed in closely related diagnostic categories and treated similarly. Until recently, evidence for shared genetic etiology derived from classical research strategies--coaggregation in family and twin studies. Accumulating sufficient numbers of families was often problematic. However, in the era of genome-wide genotyping, we can now directly estimate the degree of sharing of genetic risk factors between disorders. This strategy is practical even for very rare disorders, where it is infeasible to ascertain informative families. Importantly, the estimates of genetic correlations from genome-wide genotypes are derived using such distant relatives that contamination by shared environmental factors seems unlikely. However, any method that seeks to quantify the shared etiology of disorders assumes they can be distinguished diagnostically from one another without error. Here we investigate the impact of misdiagnosis on estimates of genetic correlation both from traditional family data and from genome-wide genotypes of case-control samples from unrelated individuals. Our analyses show similar results for levels of misdiagnosis in both types of data. In both scenarios, genetic variances and heritabilities tend to be slightly underestimated but genetic correlations are overestimated, sometimes substantially so. For example, two genetically distinct but equally heritable disorders each with prevalence 1%, can generate false-positive estimates of genetic correlations of >0.2 in the presence of 10% reciprocal misdiagnosis. Strategies for minimizing the effects of misdiagnosis in cross-disorder genetic studies are discussed.
| S-EPMC3355255 | biostudies-literature
Project description:OBJECTIVES:The ability to identify novel risk factors for health outcomes is a key strength of electronic health record (EHR)-based research. However, the validity of such studies is limited by error in EHR-derived phenotypes. The objective of this study was to develop a novel procedure for reducing bias in estimated associations between risk factors and phenotypes in EHR data. MATERIALS AND METHODS:The proposed method combines the strengths of a gold-standard phenotype obtained through manual chart review for a small validation set of patients and an automatically-derived phenotype that is available for all patients but is potentially error-prone (hereafter referred to as the algorithm-derived phenotype). An augmented estimator of associations is obtained by optimally combining these 2 phenotypes. We conducted simulation studies to evaluate the performance of the augmented estimator and conducted an analysis of risk factors for second breast cancer events using data on a cohort from Kaiser Permanente Washington. RESULTS:The proposed method was shown to reduce bias relative to an estimator using only the algorithm-derived phenotype and reduce variance compared to an estimator using only the validation data. DISCUSSION:Our simulation studies and real data application demonstrate that, compared to the estimator using validation data only, the augmented estimator has lower variance (ie, higher statistical efficiency). Compared to the estimator using error-prone EHR-derived phenotypes, the augmented estimator has smaller bias. CONCLUSIONS:The proposed estimator can effectively combine an error-prone phenotype with gold-standard data from a limited chart review in order to improve analyses of risk factors using EHR data.
| S-EPMC7025368 | biostudies-literature
Project description:In high throughput applications, such as those found in bioinformatics and finance, it is important to determine accurate probability distribution functions despite only minimal information about data characteristics, and without using human subjectivity. Such an automated process for univariate data is implemented to achieve this goal by merging the maximum entropy method with single order statistics and maximum likelihood. The only required properties of the random variables are that they are continuous and that they are, or can be approximated as, independent and identically distributed. A quasi-log-likelihood function based on single order statistics for sampled uniform random data is used to empirically construct a sample size invariant universal scoring function. Then a probability density estimate is determined by iteratively improving trial cumulative distribution functions, where better estimates are quantified by the scoring function that identifies atypical fluctuations. This criterion resists under and over fitting data as an alternative to employing the Bayesian or Akaike information criterion. Multiple estimates for the probability density reflect uncertainties due to statistical fluctuations in random samples. Scaled quantile residual plots are also introduced as an effective diagnostic to visualize the quality of the estimated probability densities. Benchmark tests show that estimates for the probability density function (PDF) converge to the true PDF as sample size increases on particularly difficult test probability densities that include cases with discontinuities, multi-resolution scales, heavy tails, and singularities. These results indicate the method has general applicability for high throughput statistical inference.
| S-EPMC5947915 | biostudies-literature
Project description:Background and objectivesPrognosis of pancreatic neuroendocrine neoplasms (PanNENs) mostly depend on tumor stage and grade, determined by Ki-67 labeling index. EUS-FNA is considered the gold-standard technique to obtain it. The aims of our study were to establish diagnostic accuracy of preoperative EUS-FNA Ki-67 evaluation considering final pathological assessment on surgical specimen as gold standard and to investigate the possible impact on prognosis of misclassification.MethodsThis is a retrospective study from a prospectively collected database. EUS-FNA grading (eG) and surgical one (sG) measured according to Ki-67 proliferative index values, according to 2017 WHO classification, were compared. eG-sG correlation was evaluated by Spearman index. Logistic regression investigated factors associated with misclassification. Prognostic difference in terms of progression-free survival was evaluated by Kaplan Meier method.ResultsOne hundred and twelve PanNENs patients enrolled. In 13.4% of patients (15/112) EUS-FNA "undergraded" patients (eG1 vs. sG2), while in 12.5% (n = 14) it "overgraded" PanNENs (eG2 to sG1). No misclassifications in G3 patients. In patients with tumors <20 mm (n = 44), 2 (4.5%) eG1 and 10 (22.7%) eG2 were finally classified respectively as G2 and G1 at surgical histology. No factors, as i.e. the lesions' size or their morphological aspect, were associated with misclassification. In overgraded PanNENs, no progression occurred, while in patients correctly classified/undergraded the progression rate was 14.3%.ConclusionsThis is the largest cohort of surgical PanNENs with preoperative EUS-FNA grading evaluation. Despite an acceptable eG-sG correlation, about 25% of patients are misclassified. Clinical impact of misclassification should be carefully considered especially in small tumors undergoing observation.
| S-EPMC8544016 | biostudies-literature
Project description:We study estimation of multivariate densities p of the form p(x) = h(g(x)) for x ? ?(d) and for a fixed monotone function h and an unknown convex function g. The canonical example is h(y) = e(-y) for y ? ?; in this case, the resulting class of densities [Formula: see text]is well known as the class of log-concave densities. Other functions h allow for classes of densities with heavier tails than the log-concave class.We first investigate when the maximum likelihood estimator p? exists for the class P(h) for various choices of monotone transformations h, including decreasing and increasing functions h. The resulting models for increasing transformations h extend the classes of log-convex densities studied previously in the econometrics literature, corresponding to h(y) = exp(y).We then establish consistency of the maximum likelihood estimator for fairly general functions h, including the log-concave class P(e(-y)) and many others. In a final section, we provide asymptotic minimax lower bounds for the estimation of p and its vector of derivatives at a fixed point x(0) under natural smoothness hypotheses on h and g. The proofs rely heavily on results from convex analysis.
| S-EPMC3058562 | biostudies-literature
Project description:In the modeling of longitudinal data from several groups, appropriate handling of the dependence structure is of central importance. Standard methods include specifying a single covariance matrix for all groups or independently estimating the covariance matrix for each group without regard to the others, but when these model assumptions are incorrect, these techniques can lead to biased mean effects or loss of efficiency, respectively. Thus, it is desirable to develop methods to simultaneously estimate the covariance matrix for each group that will borrow strength across groups in a way that is ultimately informed by the data. In addition, for several groups with covariance matrices of even medium dimension, it is difficult to manually select a single best parametric model among the huge number of possibilities given by incorporating structural zeros and/or commonality of individual parameters across groups. In this paper we develop a family of nonparametric priors using the matrix stick-breaking process of Dunson et al. (2008) that seeks to accomplish this task by parameterizing the covariance matrices in terms of the parameters of their modified Cholesky decomposition (Pourahmadi, 1999). We establish some theoretic properties of these priors, examine their effectiveness via a simulation study, and illustrate the priors using data from a longitudinal clinical trial.
| S-EPMC3852937 | biostudies-literature
Project description:BackgroundResults of bias analyses for exposure misclassification are dependent on assumptions made during analysis. We describe how adjustment for misclassification is affected by incorrect assumptions about whether sensitivity and specificity are the same (nondifferential) or different (differential) for cases and noncases.MethodsWe adjusted for exposure misclassification using probabilistic bias analysis, under correct and incorrect assumptions about whether exposure misclassification was differential or not. First, we used simulated data sets in which nondifferential and differential misclassification were introduced. Then, we used data on obesity and diabetes from the National Health and Nutrition Examination Survey (NHANES) in which both self-reported (misclassified) and measured (true) obesity were available, using literature estimates of sensitivity and specificity to adjust for bias. The ratio of odds ratio (ROR; observed odds ratio divided by true odds ratio) was used to quantify magnitude of bias, with ROR = 1 signifying no bias.ResultsIn the simulated data sets, under incorrect assumptions (eg, assuming nondifferential misclassification when it was truly differential), results were biased, with RORs ranging from 0.18 to 2.46. In NHANES, results adjusted based on incorrect assumptions also produced biased results, with RORs ranging from 1.26 to 1.55; results were more biased when making these adjustments than when using the misclassified exposure values (ROR = 0.91).ConclusionsMaking an incorrect assumption about nondifferential or differential exposure misclassification in bias analyses can lead to more biased results than if no adjustment is performed. In our analyses, incorporating uncertainty using probabilistic bias analysis was not sufficient to overcome this problem.
| S-EPMC4477528 | biostudies-literature
Project description:INTRODUCTION. Liquid biopsies are a minimally invasive collection of a patient body fluid sample. In oncology, they offer several advantages compared to traditional tissue biopsies. However, potential of this method in endometrial cancer (EC) remains poorly explored. We studied the utility of tumor educated platelets (TEPs) and circulating tumor DNA (ctDNA) for preoperative EC diagnosis, including histology determination. MATERIALS AND METHODS. TEPs from 297 subjects (53 EC patients, 40 patients with benign gynecologic conditions and 204 healthy women) were RNA-sequenced. DNA sequencing was performed in 519 primary tumor tissue samples and in16 plasma samples. Artificial intelligence was applied to sample classification. RESULTS. Platelet-dedicated classifier yielded AUC of 93.1% in test set when discriminating between healthy subjects and cancer patients. However, the discrimination between endometrial cancer and benign gynecologic conditions was relatively low, with AUC of 60.7%. ctDNA-dedicated classifier discriminated primary tumor tissue samples with AUC of 91.4% and ctDNA blood samples with AUC of 87.5%. CONCLUSIONS Liquid biopsies show potential in EC diagnosis. Both TEPs and ctDNA profiles coupled with artificial intelligence constitute a source of useful information. Further work, involving more cases, is warranted.
2021-12-01 | GSE184904 | GEO
Project description:ObjectivesTo assess the methodologies used in the estimation of diagnostic accuracy of SARS-CoV-2 real-time reverse transcription polymerase chain reaction (rRT-PCR) and other nucleic acid amplification tests (NAATs) and to evaluate the quality and reliability of the studies employing those methods.MethodsWe conducted a systematic search of English-language articles published December 31, 2019-June 19, 2020. Studies of any design that performed tests on ≥10 patients and reported or inferred correlative statistics were included. Studies were evaluated using elements of the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) guidelines.ResultsWe conducted a narrative and tabular synthesis of studies organized by their reference standard strategy or comparative agreement method, resulting in six categorizations. Critical study details were frequently unreported, including the mechanism for patient/sample selection and researcher blinding to results, which lead to concern for bias.ConclusionsCurrent studies estimating test performance characteristics have imperfect study design and statistical methods for the estimation of test performance characteristics of SARS-CoV-2 tests. The included studies employ heterogeneous methods and overall have an increased risk of bias. Employing standardized guidelines for study designs and statistical methods will improve the process for developing and validating rRT-PCR and NAAT for the diagnosis of COVID-19.
| S-EPMC7457918 | biostudies-literature