Methods of Estimating Correlation Coefficients in the Presence of Influential Outlier(s)

Correlation methods are indispensable in the study of the linear relationship between two variables. However, many researchers often adopt inappropriate correlation methods in the study of linear relationships which usually leads to unreliable results. Recurrently, most researchers ignorantly employ the Pearson method in a dataset that contained outliers, instead of more appropriate correlation methods such as Spearman, Kendall Tau, Median and Quadrant which might be suitable in the calculation of correlation coefficient in the presence of influential outliers. It is noted that the accuracy of estimation of correlation coefficients under outliers has been a long-standing problem for methodological researchers. This is due to low knowledge of correlation methods and their assumptions which have led to inappropriate application of correlation methods in research analysis. Five different methods of estimating correlation coefficients in the presence of influential outlier (contaminated data) were considered: Pearson Correlation Coefficient, Spearman Correlation Coefficient, Kendall Tau Correlation Coefficient, Median Correlation Coefficient and Quadrant Correlation Coefficient.

Weighted Maximum Likelihood Correlation Coefficient to Handle Missing Values and Outliers in Data Set

The proposed estimator, namely weighted maximum likelihood (WML) correlation coefficient, for measuring the relationship between two variables to concern about missing values and outliers in the dataset is presented. This estimator is proven by applying the conditional probability function to take care of some missing values and pay more attention to values near the center. However, outliers in the dataset are assigned a slight weight. These using techniques will give the robust proposed method when the preliminary assumptions are not met data analysis. To inspect about the quality of the proposed estimator, the six methods—WML, Pearson, median, percentage bend, biweight mid, and composite correlation coefficients—are compared the properties in two criteria, i.e. the bias and mean squared error, via the simulation study. The results of generated data are illustrated that the WML estimator seems to have the best performance to withstand the missing values and outliers in dataset, especially for the tiny sample size and large percentage of outliers regardless of missing data levels. However, for the massive sample size, the median correlation coefficient seems to have the good estimator when linear relationship levels between two variables are approximately over 0.4 irrespective of outliers and missing data levels

An integrated landscape of protein expression in human cancer

Using 11 proteomics datasets, mostly available through the PRIDE database, we assembled a reference expression map for 191 cancer cell lines and 246 clinical tumour samples, across 13 lineages. We found unique peptides identified only in tumour samples despite a much higher coverage in cell lines. These were mainly mapped to proteins related to regulation of signalling receptor activity. Correlations between baseline expression in cell lines and tumours were calculated. We found these to be highly similar across all samples with most similarity found within a given sample type. Integration of proteomics and transcriptomics data showed median correlation across cell lines to be 0.58 (range between 0.43 and 0.66). Additionally, in agreement with previous studies, variation in mRNA levels was often a poor predictor of changes in protein abundance. To our knowledge, this work constitutes the first meta-analysis focusing on cancer-related public proteomics datasets. We therefore also highlight shortcomings and limitations of such studies. All data is available through PRIDE dataset identifier PXD013455 and in Expression Atlas.

Three dimensional reconstruction to visualize atrial fibrillation activation patterns on curved atrial geometry

Background The rotational activation created by spiral waves may be a mechanism for atrial fibrillation (AF), yet it is unclear how activation patterns obtained from endocardial baskets are influenced by the 3D geometric curvature of the atrium or ‘unfolding’ into 2D maps. We develop algorithms that can visualize spiral waves and their tip locations on curved atrial geometries. We use these algorithms to quantify differences in AF maps and spiral tip locations between 3D basket reconstructions, projection onto 3D anatomical shells and unfolded 2D surfaces. Methods We tested our algorithms in N = 20 patients in whom AF was recorded from 64-pole baskets (Abbott, CA). Phase maps were generated by non-proprietary software to identify the tips of spiral waves, indicated by phase singularities. The number and density of spiral tips were compared in patient-specific 3D shells constructed from the basket, as well as 3D maps from clinical electroanatomic mapping systems and 2D maps. Results Patients (59.4±12.7 yrs, 60% M) showed 1.7±0.8 phase singularities/patient, in whom ablation terminated AF in 11/20 patients (55%). There was no difference in the location of phase singularities, between 3D curved surfaces and 2D unfolded surfaces, with a median correlation coefficient between phase singularity density maps of 0.985 (0.978–0.990). No significant impact was noted by phase singularities location in more curved regions or relative to the basket location (p>0.1). Conclusions AF maps and phase singularities mapped by endocardial baskets are qualitatively and quantitatively similar whether calculated by 3D phase maps on patient-specific curved atrial geometries or in 2D. Phase maps on patient-specific geometries may be easier to interpret relative to critical structures for ablation planning.

The pons as reference region for intensity normalization in semi-quantitative analysis of brain 18FDG PET: application to metabolic changes related to ageing in conventional and digital control databases

Background The objective of the study is to define the most appropriate region for intensity normalization in brain 18FDG PET semi-quantitative analysis. The best option could be based on previous absolute quantification studies, which showed that the metabolic changes related to ageing affect the quasi-totality of brain regions in healthy subjects. Consequently, brain metabolic changes related to ageing were evaluated in two populations of healthy controls who underwent conventional (n = 56) or digital (n = 78) 18FDG PET/CT. The median correlation coefficients between age and the metabolism of each 120 atlas brain region were reported for 120 distinct intensity normalizations (according to the 120 regions). SPM linear regression analyses with age were performed on most significant normalizations (FWE, p < 0.05). Results The cerebellum and pons were the two sole regions showing median coefficients of correlation with age less than − 0.5. With SPM, the intensity normalization by the pons provided at least 1.7- and 2.5-fold more significant cluster volumes than other normalizations for conventional and digital PET, respectively. Conclusions The pons is the most appropriate area for brain 18FDG PET intensity normalization for examining the metabolic changes through ageing.

The Pons As Region of Reference For Intensity Normalization In Semi-Quantitative Analysis of Brain 18FDG PET: Application to Metabolic Changes Related to Ageing in Conventional and Digital Control Databases

Background: To define the most appropriate region for intensity normalization in brain 18FDG PET analysis through ageing.Brain metabolic changes related to ageing were evaluated in two populations of healthy controls who underwent conventional (n=56) or digital (n=78) 18FDG PET/CT. The median correlation coefficients between age and the metabolism of each 120 atlas brain region were reported for 120 distinct intensity normalizations (according to the 120 regions). SPM linear regression analyses with age were performed on most significant normalizations (FWE, p<0.05).Results: The cerebellum and pons were the two sole regions showing median coefficients of correlation with age less than -0.5. With SPM, the intensity normalization through the pons provided at least 1.7- and 2.5-fold more significant cluster volumes than other normalizations for conventional and digital PET respectively. Conclusions: The pons is the most appropriate area for brain 18FDG PET intensity normalization for examining the metabolic changes through ageing.

On the usage of average Hausdorff distance for segmentation performance assessment: hidden error when used for ranking

Average Hausdorff distance is a widely used performance measure to calculate the distance between two point sets. In medical image segmentation, it is used to compare ground truth images with segmentations allowing their ranking. We identified, however, ranking errors of average Hausdorff distance making it less suitable for applications in segmentation performance assessment. To mitigate this error, we present a modified calculation of this performance measure that we have coined “balanced average Hausdorff distance”. To simulate segmentations for ranking, we manually created non-overlapping segmentation errors common in magnetic resonance angiography cerebral vessel segmentation as our use-case. Adding the created errors consecutively and randomly to the ground truth, we created sets of simulated segmentations with increasing number of errors. Each set of simulated segmentations was ranked using both performance measures. We calculated the Kendall rank correlation coefficient between the segmentation ranking and the number of errors in each simulated segmentation. The rankings produced by balanced average Hausdorff distance had a significantly higher median correlation (1.00) than those by average Hausdorff distance (0.89). In 200 total rankings, the former misranked 52 whilst the latter misranked 179 segmentations. Balanced average Hausdorff distance is more suitable for rankings and quality assessment of segmentations than average Hausdorff distance.

The Estimates Item Parameter for Multidimensional Three-Parameter Logistics

The purpose of this study is to measure the accuracy of item parameters and abilities by using the Multidimensional Three-Parameter Logistics (M3PL) model. M3PL is a series of tests that measure more than one dimension of ability (θ). Item parameter estimation and the ability to model M3PL are reviewed based on a sample size of 1000 and test lengths of 15, 25, and 40. Parameter estimations are obtained using the Wingen software that is converted to BILOG. The results show that the estimate obtained with a test length of 15 displays a median correlation of 0.787 (high). The study therefore concludes that the level of difficulty of the questions is higher or the questions given to respondents are more difficult, so many respondents guessed the answers. The results of the estimated grain parameters and capabilities indicated that scoring based on sample size greatly affects the stability of the test length. By using the M3PL model, parameters can be measured pseudo-guessing, parameters b and parameters a. MIRT is able to explain interactions between the items on the test and the answers of the participants. The estimated results of the item parameters and the ability parameters of the participants also proved to be accurate and efficient. Keywords: Multidimensional Three-Parameter Logistics (M3PL), distribution parameter, test length

Estimating regional pulmonary blood flow in EIT with regularized deconvolution with a Tikhonov regularization

Electrical Impedance Tomography (EIT) is a clinically used tool for bed-side monitoring of ventilation. Previous work also showed a high potential for lung perfusion monitoring with indicator-enhanced EIT. However, many research questions have yet to be answered before it can be broadly applied in clinical everyday life. The goal of this work is to evaluate a new method to improve EIT perfusion measurements. Pulmonary hemodynamic transfer functions were estimated using regularized deconvolution with Tikhonov regularization to estimate spatial perfusion parameters. The final comparison between EIT images and PET scans showed a median correlation of 0.897 for the images which were reconstructed using the regularized deconvolution. In comparison the previously used maximum slope method led to a median correlation of 0.868.

Development, validation and utilisation of dish-based dietary assessment tools: a scoping review

Objective: To summarise the existing evidence of development, validation and current status of utilisation of dish-based dietary assessment tools. Design: Scoping review. Setting: Systematic search using PubMed and Web of Science. Results: We identified twelve tools from seventy-four eligible publications. They were developed for Koreans (n 4), Bangladeshis (n 2), Iranians (n 1), Indians/Malays/Chinese (n 1), Japanese (n 3) and Chinese Americans (n 1). Most tools (10/12) were composed of a dish-based FFQ. Although the development process of a dish list varied among the tools, six studies classified mixed dishes based on the similarity of their characteristics such as food ingredients and cooking methods. Tools were validated against self-reported dietary information (n 9) and concentration biomarkers (n 1). In the eight studies assessing the differences between the tool and a reference, the mean (or median) intake of energy significantly differed in five studies, and 26–83 % of nutrients significantly differed in eight studies. Correlation coefficients for energy ranged from 0·15 to 0·87 across the thirteen studies, and the median correlation coefficients for nutrients ranged from 0·12 to 0·77. Dish-based dietary assessment tools were used in fifty-nine studies mainly to assess diet–disease relationships in target populations. Conclusions: Dish-based dietary assessment tools have exclusively been developed and used for Asian-origin populations. Further validation studies, particularly biomarker-based studies, are needed to assess the applicability of tools.