arXiv subjects

Fatma Parlak

Publications and source records attributed to Fatma Parlak.

3 recordsLinked to original sources

arXiv2025

Robust normality transformation for outlier detection in diverse distributions, with application to functional neuroimaging data

Automatic detection of statistical outliers is facilitated through knowledge of the source distribution of regular observations. Since the population distribution is often unknown in practice, one approach is to apply a transformation to Normality. However, the efficacy of transformation is hindered by the presence of outliers, which can have an outsized influence on transformation parameter(s) and lead to masking of outliers post-transformation. Robust Box-Cox and Yeo-Johnson transformations have been proposed but those transformations are only equipped to deal with skew. Here, we develop a novel robust method for transformation to Normality based on the highly flexible sinh-arcsinh (SHASH) family of distributions, which can accommodate skew, non-Gaussian tail weights, and combinations of both. A critical step is initializing outliers, given their potential influence on the highly flexible SHASH transformation. To this end, we consider conventional robust z-scoring and a novel anomaly detection approach. Through extensive simulation studies and real data analyses representing a wide variety of distribution shapes, we find that SHASH transformation outperforms existing methods, exhibiting high sensitivity to outliers even at heavy contamination levels (20-30\%). We illustrate the utility of SHASH transformation-based outlier detection in the context of noise reduction in functional neuroimaging data.

stat.ME↗

arXiv2023

A robust multivariate, non-parametric outlier identification method for scrubbing in fMRI

Functional magnetic resonance imaging (fMRI) data contain high levels of noise and artifacts. To avoid contamination of downstream analyses, fMRI-based studies must identify and remove these noise sources prior to statistical analysis. One common approach is the "scrubbing" of fMRI volumes that are thought to contain high levels of noise. However, existing scrubbing techniques are based on ad hoc measures of signal change. We consider scrubbing via outlier detection, where volumes containing artifacts are considered multidimensional outliers. Robust multivariate outlier detection methods are proposed using robust distances (RDs), which are related to the Mahalanobis distance. These RDs have a known distribution when the data are i.i.d. normal, and that distribution can be used to determine a threshold for outliers where fMRI data violate these assumptions. Here, we develop a robust multivariate outlier detection method that is applicable to non-normal data. The objective is to obtain threshold values to flag outlying volumes based on their RDs. We propose two threshold candidates that embark on the same two steps, but the choice of which depends on a researcher's purpose. Our main steps are dimension reduction and selection, robust univariate outlier imputation to get rid of the effect of outliers on the distribution, and estimating an outlier threshold based on the upper quantile of the RD distribution without outliers. The first threshold candidate is an upper quantile of the empirical distribution of RDs obtained from the imputed data. The second threshold candidate calculates the upper quantile of the RD distribution that a nonparametric bootstrap uses to account for uncertainty in the empirical quantile. We compare our proposed fMRI scrubbing method to motion scrubbing, data-driven scrubbing, and restrictive parametric multivariate outlier detection methods.

stat.ME↗

arXiv2022

Sources of residual autocorrelation in multiband task fMRI and strategies for effective mitigation

In task fMRI analysis, OLS is typically used to estimate task-induced activation in the brain. Since task fMRI residuals often exhibit temporal autocorrelation, it is common practice to perform prewhitening prior to OLS to satisfy the assumption of residual independence, equivalent to GLS. While theoretically straightforward, a major challenge in prewhitening in fMRI is accurately estimating the residual autocorrelation at each location of the brain. Assuming a global autocorrelation model, as in several fMRI software programs, may under- or over-whiten particular regions and fail to achieve nominal false positive control across the brain. Faster multiband acquisitions require more sophisticated models to capture autocorrelation, making prewhitening more difficult. These issues are becoming more critical now because of a trend towards subject-level analysis, where prewhitening has a greater impact than in group-average analyses. In this article, we first thoroughly examine the sources of residual autocorrelation in multiband task fMRI. We find that residual autocorrelation varies spatially throughout the cortex and is affected by the task, the acquisition method, modeling choices, and individual differences. Second, we evaluate the ability of different AR-based prewhitening strategies to effectively mitigate autocorrelation and control false positives. We find that allowing the prewhitening filter to vary spatially is the most important factor for successful prewhitening, even more so than increasing AR model order. To overcome the computational challenge associated with spatially variable prewhitening, we developed a computationally efficient R implementation based on parallelization and fast C++ backend code. This implementation is included in the open source R package BayesfMRI.

stat.ME↗