ANOVA-based Method Research Articles

Fatigue response of AlSi10Mg by laser powder bed fusion: influence of build orientation, heat, and surface treatments

AbstractThe aim of this study is to analyze how the fatigue behavior of AlSi10Mg by laser powder bed fusion is affected by build orientation, heat, and surface treatments. A three-by-three factorial plan has been arranged for this purpose. Particularly, regarding the heat treatment, three levels were considered (as built, age hardening, and stress relief); whereas, for the surface treatment, three levels were investigated (micro-shot-peening, micro-shot-peening plus fine blasting, and machining and lapping following laser powder bed fusion). Regarding the build orientation, the specimens were manufactured using three different build orientations (0°, 45°, and 90°). The obtained data have been statistically analyzed by a three-factor ANOVA-based method. The results, supported by fractographic and micrographic microscopy analyses, indicate that the age-hardening treatment yields the maximum benefits, whereas stress relief may even have a detrimental effect. As for surface treatments, a positive influence of shot-peening has been found.

Comparison of Multivariate ANOVA-Based Approaches for the Determination of Relevant Variables in Experimentally Designed Metabolomic Studies

The use of chemometric methods based on the analysis of variances (ANOVA) allows evaluation of the statistical significance of the experimental factors used in a study. However, classical multivariate ANOVA (MANOVA) has a number of requirements that make it impractical for dealing with metabolomics data. For this reason, in recent years, different options have appeared that overcome these limitations. In this work, we evaluate the performance of three of these multivariate ANOVA-based methods (ANOVA simultaneous component analysis—ASCA, regularized MANOVA–rMANOVA, and Group-wise ANOVA-simultaneous component analysis—GASCA) in the framework of metabolomics studies. Our main goals are to compare these various ANOVA-based approaches and evaluate their performance on experimentally designed metabolomic studies to find the significant factors and identify the most relevant variables (potential markers) from the obtained results. Two experimental data sets were generated employing liquid chromatography coupled to mass spectrometry (LC-MS) with different complexity in the design to evaluate the performance of the statistical approaches. Results show that the three considered ANOVA-based methods have a similar performance in detecting statistically significant factors. However, relevant variables pointed by GASCA seem to be more reliable as there is a strong similarity with those variables detected by the widely used partial least squares discriminant analysis (PLS-DA) method.

Filling the matrix: an ANOVA-based method to emulate regional climate model simulations for equally-weighted properties of ensembles of opportunity

Collections of large ensembles of regional climate model (RCM) downscaled climate data for particular regions and scenarios can be organized in a usually incomplete matrix consisting of GCM (global climate model) x RCM combinations. When simple ensemble averages are calculated, each GCM will effectively be weighted by the number of times it has been downscaled. In order to facilitate more equal and less arbitrary weighting among downscaled GCM results, we present a method to emulate the missing combinations in such a matrix, enabling equal weighting among participating GCMs and hence among regional consequences of large-scale climate change simulated by each GCM. This method is based on a traditional Analysis of Variance (ANOVA) approach. The method is applied and studied for fields of seasonal average temperature, precipitation and surface wind and for the 10-year return value of daily precipitation and of 10-m wind speed for a completely filled matrix consisting of 5 GCMs and 4 RCMs. We quantify the skill of the two averaging methods for different numbers of missing simulations and show that ensembles where lacking members have been emulated by the ANOVA technique are better at representing the full ensemble than corresponding simple ensemble averages, particularly in cases where only a few model combinations are absent. The technique breaks down when the number of missing simulations reaches the sum of the numbers of GCMs and RCMs. Also, the method is only useful when inter-simulation variability is limited. This is the case for the average fields that have been studied, but not for the extremes. We have developed analytical expressions for the degree of improvement obtained with the present method, which quantify this conclusion.

Observer Variability: A New Approach in Evaluating Interobserver Agreement

Existing indices of observer agreement for continuous data, such as the intraclass correlation coefficient or the concordance correlation coefficient, measure the total observer-related variability, which includes the variabilities between and within observers. This work introduces a new index that measures the interobserver variability, which is defined in terms of the distances among the ‘true values’ assigned by different observers on the same subject. The new coefficient of interobserver variability (CIV ) is defined as the ratio of the interobserver and the total observer variability. We show how to estimate the CIV and how to use bootstrap and ANOVAbased methods for inference. We also develop a coefficient of excess observer variability, which compares the total observer variability to the expected total observer variability when there are no differences among the observers. This coefficient is a simple function of the CIV . In addition, we show how the value of the CIV , estimated from an agreement study, can be used in the design of measurements studies. We illustrate the new concepts and methods by two examples, where (1) two radiologists used calcium scores to evaluate the severity of coronary artery arteriosclerosis, and (2) two methods were used to measure knee joint angle.

Abstract 535: Mathematical Modeling of Hyperpolarized Pyruvate Metabolism in Human Heart

Background: Dissolution dynamic nuclear polarization is a novel method that increases more than 10,000-fold the 13 C signal-to-noise ratio and emerges as a useful tool for detection of molecules much less prevalent than water in the human body. Hyperpolarized [1- 13 C] pyruvate (HP) permits the noninvasive, nonradioactive study of metabolic flux via pyruvate dehydrogenase, a key enzyme in the complex cardiac adaptation to physical activity, energy sources, and in various disease states. The aim of this study is to define an analysis pipeline of multiparametric, non-steady-state, real-time human cardiac magnetic resonance spectroscopic investigations. Methods: HP was infused intravenously into five human subjects. Dynamic spectroscopic data was acquired for [1- 13 C]- pyruvate, 13 C-alanine, [1- 13 C]-lactate, and 13 C-bicarbonate at two consecutive times for each subject. Imaging data were exported into MATLAB. We employed a first-order kinetic model to fit the signal for the hyperpolarized metabolites assuming bidirectional flow between pyruvate and lactate as well as pyruvate and alanine. A gamma variate input function was used to model the initial pyruvate bolus. We also calculated non-parametric values to describe the signal for each metabolite, specifically area-under-curve (AUC) and time-to-peak (TTP). The repeatability analysis was performed using an ANOVA-based method. Results: Rate constants derived from the kinetic model: kPL 0.016 ± 0.008s -1 , kPA 0.012 ± 0.006s -1 , kPB 0.024±0.0095s -1 are of similar order of magnitude to prior studies. The reverse constant for lactate to pyruvate conversion (kLP) is relatively unchanged across the subjects (0.0099±0.0005), and reverse constant for alanine to pyruvate (kAP) conversion is almost negligible (0.0008 ± 0.0012). The model well describes the metabolites signal over time. Within subject repeatability was higher for TTP (lactate: 0.981 ± 0.018) ) than for rate constants (kPL 0.823 ± 0.153) and AUC (lactate 0.697 ± 0.244) with kPB having the lowest repeatability (0.325 ± 0.424) due to a single outlier. Conclusions: We present a real-time pipeline for analysis and modelling of pyruvate metabolism in the human heart and confirm that it is reproducible within the same subject under same conditions.

Machine Learning with Imbalanced EEG Datasets using Outlier-based Sampling.

Epilepsy is a neurological disorder which causes seizures in over 65 million people worldwide. Recently developed implantable therapeutic devices aim to prevent symptoms by applying acute electrical stimulation to the seizure-generating brain region in response to activity detected by on-device machine learning hardware. Many training algorithms require an equal number of examples for each target class (e.g. normal activity and seizures), and performance can suffer if this condition is not satisfied. In the case of epilepsy, poor performance can cause seizures to be missed, or stimulation to be applied erroneously. As there is an abundance of normal (interictal) data in clinical EEG recordings, but seizures are rare events (less than 1% of the dataset), the data available for training is severely imbalanced. There are several conventional pre-processing methods used to address imbalanced class learning, such as down-sampling of the majority class and up-sampling of the minority class, but each have performance drawbacks. This paper presents an improved method which involves reducing the majority class down to the most effective interictal outlier samples. Outliers are determined by using Exponentially Decaying Memory Signal Energy (EDMSE) features with Isolation Forests and an ANOVA-based method, which involves comparing a moving feature window to a baseline reference window. Outlier-based sampling is tested with two classifiers (KNN and Logistic Regression) and achieves higher accuracy (∼2% increase) and fewer false positives (∼38% decrease), along with a lower latency (∼3 seconds shorter) compared to conventional training set pre-processing methods.

Uticaj formulacije tableta sa trenutnim oslobađanjem na profil brzine rastvaranja paracetamola

Introduction: Paracetamol is analgesic and antipyretic, which is usually in the form of an immediate release tablet formulations. Therapeutic effects in terms of the speed and intensity of the analgesic effect is dependent on speed of liberation from formulation. Aim: The aim of this work was to determine and compare dissolution profiles of 4 paracetamol immediate release tablet formulations and to determine influence of excipients on kinetic of paracetamol dissolution. Materials and Methods: Dissolution profiles of paracetamol tablets were determined using method with paddles and phosphate buffer pH 6.8 as a medium. Release of paracetamol was followed 60 minutes (using 6 time points). Concentration of paracetamol was measured using UV/Vis spectrophotometric method (243 nm). Dissolution profiles were compared using model-independent method (difference factor and similarity factor), statistic method (ANOVA-based method and pair Student's T-test, p<0.05) and model dependent methods (to determinate the release kinetics of paracetamol). Results: All formulations in the first 45 minutes liberated more than 85 % of the labled content. Formulation D, which contained superdesintegrator, released 90% of the content in the first 5 minutes. Though based on values of difference and similarity factors formulations are, not significantly different, ANOVA-based method showed that formulation A and B, B and C, as well as formulation B and D do statistically differ in all 6 time points, meaning they have parallel profiles. The release of paracetamol from formulations A and D is best described by the first order kinetic model, while the release of formulations B and C by the logistic model. Conclusions: The release kinetic of paracetamol is mostly influenced by the type of superdisintegrants and lubricants. Formulation with superdisintegrant technology OptiZorb® demonstrated fastes release rate and thus it is expected to produce the fastest pharmacodynamic effect. .

Restricted Maximum Likelihood Estimation for Parameters of the Social Relations Model.

In many areas of research, the round-robin design is used to study interpersonal judgments and behaviors. The resulting data are analyzed with the social relations model (SRM), whereby almost all previously published studies have used ANOVA-based methods or multilevel-based methods to obtain SRM parameter estimates. In this article, the SRM is embedded into the linear mixed model framework, and it is shown how restricted maximum likelihood can be employed to estimate the SRM parameters. It is also described how the effect of covariates on the SRM-specific effects can be estimated. An example is presented to illustrate the approach. We also present the results of a simulation study in which the performance of the proposed approach is compared to the ANOVA method.

Identification of suitable grapevine reference genes for qRT-PCR derived from heterologous species.

Identification and validation of suitable reference genes that exhibit robust transcriptional stability across many sample types is an absolute requirement of all qRT-PCR experiments. Often, however, only small numbers of reference genes, validated across limited sample types, are available for non-model species. This points to a clear need to assess and validate a wider range of potential reference genes than is currently available. We therefore looked to test and validate a large number of potential reference genes across a wide range of tissue types and treatments to determine the applicability of these reference genes for use in grapevine and other non-model plant species. Potential reference genes were selected based on stability of gene transcription in the model plant species Arabidopsis or due to their common use in the grapevine community. The selected reference genes were analyzed across two datasets consisting of a range of either 'Sauvignon blanc' or 'Pinot noir' tissues. A total of 11 potential reference genes were screened across the two datasets. Gene stability was analyzed by GeNorm, a widely used Excel application, or an ANOVA-based method developed in red clover. Both analysis methods showed that all 11 potential reference genes are stably expressed in the datasets tested, but the rankings of gene stability differed based on the datasets and analysis method used. Furthermore, the transcript stability of these genes, initially identified in Arabidopsis and now validated in grapevine, suggests applicability across a wide range of non-model plant species in addition to their utility in grapevine.

Quantifying polymorphism and divergence from epigenetic data: a framework for inferring the action of selection.

Epigenetic modifications are alterations that regulate gene expression without modifying the underlying DNA sequence. DNA methylation and histone modifications, for example, are capable of spatial and temporal regulation of expression—with several studies demonstrating that these epigenetic marks are heritable. Thus, like DNA sequence, epigenetic marks are capable of storing information and passing it from one generation to the next. Because the epigenome is dynamic and epigenetic modifications can respond to external environmental stimuli, such changes may play an important role in adaptive evolution. While recent studies provide strong evidence for species-specific signatures of epigenetic marks, little is known about the mechanisms by which such modifications evolve. In order to address this question, we analyze the genome wide distribution of an epigenetic histone mark (H3K4me3) in prefrontal cortex neurons of humans, chimps and rhesus macaques. We develop a novel statistical framework to quantify within- and between-species variation in histone methylation patterns, using an ANOVA-based method and defining an FST -like measure for epigenetics (termed epi- FST), in order to develop a deeper understanding of the evolutionary pressures acting on epigenetic variation. Results demonstrate that genes with high epigenetic FST values are indeed significantly overrepresented among genes that are differentially expressed between species, and we observe only a weak correlation with SNP density.

Robust Means Modeling

This study proposes robust means modeling (RMM) approaches for hypothesis testing of mean differences for between-subjects designs in order to control the biasing effects of nonnormality and variance inequality. Drawing from structural equation modeling (SEM), the RMM approaches make no assumption of variance homogeneity and employ robust estimation/rescaling strategies in order to alleviate reliance on normality. A Monte Carlo simulation is conducted to compare the Type I error rate and the power of the proposed six RMM test statistics to five analysis of variance (ANOVA)-based statistics, the latter of which have also employed trimmed means and Winsorized variances to enhance robustness. Various simulation factors manipulated include variance inequality, sample-size pairings with group variances, degree of nonnormality, alpha level for hypothesis tests, and effect size. Results show that the proposed RMM methods are indeed superior to the traditional ANOVA-based methods.

A comparison of association methods for cytotoxicity mapping in pharmacogenomics.

Cytotoxicity assays of immortalized lymphoblastoid cell lines (LCLs) represent a promising new in vitro approach in pharmacogenomics research. However, previous studies employing LCLs in gene mapping have used simple association methods, which may not adequately capture the true differences in non-linear response profiles between genotypes. Two common approaches summarize each dose-response curve with either the IC50 or the slope parameter estimates from a hill slope fit and treat these estimates as the response in a linear model. The current study investigates these two methods, as well as four novel methods, and compares their power to detect differences between the response profiles of genotypes under a variety of different alternatives. The four novel methods include two methods that summarize each dose-response by its area under the curve, one method based off of an analysis of variance (ANOVA) design, and one method that compares hill slope fits for all individuals of each genotype. The power of each method was found to depend not only on the choice of alternative, but also on the choice for the set of dosages used in cytotoxicity measurements. The ANOVA-based method was found to be the most robust across alternatives and dosage sets for power in detecting differences between genotypes.

Prediction intervals for general balanced linear random models

The main interest of prediction intervals lies in the results of a future sample from a previously sampled population. In this article, we develop procedures for the prediction intervals which contain all of a fixed number of future observations for general balanced linear random models. Two methods based on the concept of a generalized pivotal quantity (GPQ) and one based on ANOVA estimators are presented. A simulation study using the balanced one-way random model is conducted to evaluate the proposed methods. It is shown that one of the two GPQ-based and the ANOVA-based methods are computationally more efficient and they also successfully maintain the simulated coverage probabilities close to the nominal confidence level. Hence, they are recommended for practical use. In addition, one example is given to illustrate the applicability of the recommended methods.

Comparison of In Vitro Dissolution Profiles of Oxcarbazepine-HP b-CD Tablet Formulations with Marketed Oxcarbazepine Tablets

The aims of this study were (1) to compare the in vitro dissolution profiles of oxcarbazepine-HP b-CD tablet formulations with those of marketed oxcarbazepine tablets, (2) to apply statistical models to evaluate each method in terms of easy application and usefulness, and (3) to identify the advantages and disadvantages of each method. The results show that the tablets containing hydroxypropyl b-cyclodextrin (HP b-CD) with sodium carboxymethylcellulose (NaCMC) exhibit faster release (1.93-fold) than marketed oxcarbazepine (OXO) tablets. From Weibull parameters, it was shown that Td is three times higher for OXC-HP b-CD with NaCMC tablets than for marketed oxcarbazepine tablets. The release kinetics of OXC complexed with HP b-CD from different tablets was investigated using several mathematical equations. Model-independent methods including difference factor, f1, and similarity factor, f2; model-dependent methods; and ANOVA-based methods were used for the comparison of in vitro dissolution profiles. The results show that ANOVA-based methods and model-dependent methods are more discriminative than model-independent methods. Modelindependent methods seem to be easier to apply and interpret; only one value is obtained to describe the closeness of the two dissolution profiles. The application and evaluation of model-dependent methods are more complicated; these methods present an acceptable model approach to the true relationship between percent dissolved and time variables, including statistical assumptions that could be checked. Drug release data fit well to the Hixson–Crowell model. The drug release mechanism was a graphic of the cube root of the unreleased fraction of the drug with time. The Weibull model was more useful for comparing the release profiles. Weibull parameters were more sensitive to the differences between the two release kinetic data sets.

Effect of uncertainty about population parameters on pharmacodynamics-based prediction of clinical trial power

Clinical trial simulation (CTS) may be applied to predict power of intended drug trials on the basis of pharmacokinetic/pharmacodynamic (PKPD) drug models. The validity of such predictions will, among other factors, depend on the degree of uncertainty about population parameters entering the simulation. In the current article, we illustrate how population parameter uncertainty may be incorporated in the overall simulation model, using a worked example to demonstrate our approach. Moreover, we suggest an ANOVA-based method for sensitivity analysis, aimed at distinguishing important population parameters, required to be input in the model with a low degree of uncertainty for precise power predictions, from unimportant parameters, which may be entered with a high degree of uncertainty without affecting the reliability of predictions. Our results from simulations with different settings of hyperparameters and doses suggest that indices derived from this type of sensitivity analysis may be used for grading the influence on prediction quality of uncertainty about different population parameters and, thus, facilitate the allocation of resources expended for the preparation of a successful CTS project.

Nonlinear mixed effects modelling for estimation of steady state attainment

Assessment of the time needed to attain steady state is a key pharmacokinetic objective during drug development. Traditional approaches for assessing steady state include ANOVA-based methods for comparing mean plasma concentration values from each sampling day, with either a difference or equivalence test. However, hypothesis-testing approaches are ill suited for assessment of steady state. This paper presents a nonlinear mixed effects modelling approach for estimation of steady state attainment, based on fitting a simple nonlinear mixed model to observed trough plasma concentrations. The simple nonlinear mixed model is developed and proposed for use under certain pharmacokinetic assumptions. The nonlinear mixed modelling estimation approach is described and illustrated by application to trough data from a multiple dose trial in healthy subjects. The performance of the nonlinear mixed modelling approach is compared to ANOVA-based approaches by means of simulation techniques. Copyright © 2005 John Wiley & Sons, Ltd.

Comparison of in vitro dissolution profiles by ANOVA-based, model-dependent and -independent methods

In this study, the aim was to apply different comparison methods to dissolution profiles of immediate release commercial film-coated tablets of naproxen sodium in order to (1) evaluate each method in terms of easy application and usefulness and (2) identify the advantages and disadvantages of each method. Dissolution testing was conducted using the USP monograph of naproxen sodium. The applied methods for the comparison of in vitro dissolution profiles are ANOVA-based methods, model-dependent methods, and model-independent methods including difference factor, f 1, and similarity factor, f 2. All the methods appear to be applicable and useful in comparing dissolution profiles. The results show that ANOVA-based methods and model-dependent methods are more discriminative than the f-factors. f-Factors seem to be easier to apply and interpret; only one value is obtained to describe the closeness of the two dissolution profiles. However, a last point for dissolution had to be determined, since the values of the f-factors depend on this point. The application and evaluation of model-dependent methods are more complicated; these methods present an acceptable model approach to the true relationship between percent dissolved and time variables, including statistical assumptions which could be checked. Dissolution profiles can be tested for differences in both level and shape by ANOVA-based methods and these methods provide detailed information about dissolution data which can be useful also in formulation development to match release to a reference product.