arXiv subjects

Simon Vandekar

Publications and source records attributed to Simon Vandekar.

8 recordsLinked to original sources

arXiv2026

Evaluation of neuroCombat and deep learning harmonization for multi-site magnetic resonance neuroimaging in youth with prenatal alcohol exposure

In cases of prevalent diseases and disorders, such as Prenatal Alcohol Exposure (PAE), multi-site data collection allows for increased study samples. However, multi-site studies introduce additional variability through heterogeneous collection materials, such as scanner and acquisition protocols, which confound with biologically relevant signals. Neuroscientists often utilize statistical methods on image-derived metrics, such as volume of regions of interest, after all image processing to minimize site-related variance. HACA3, a deep learning harmonization method, offers an opportunity to harmonize image signals prior to metric quantification; however, HACA3 has not yet been validated in a pediatric cohort. In this work, we investigate HACA3's ability to remove site-related variance and preserve biologically relevant signal compared to a statistical method, neuroCombat, and pair HACA3 processing with neuroCombat to evaluate the efficacy of multiple harmonization methods in a pediatric (age 7 to 21) population across three unique scanners with controls and cases of PAE with downstream MaCRUISE volume metrics. We find that HACA3 qualitatively improves inter-site contrast variations, but statistical methods reduce greater site-related variance within the MaCRUISE volume metrics following an ANCOVA test, and HACA3 relies on follow-up statistical methods to approach maximal biological preservation in this context.

eess.IV↗

arXiv2026

Asymptotic Distribution of Robust Effect Size Index

The Robust Effect Size Index (RESI) is a recently proposed standardized effect size to quantify association strength across models. However, its confidence interval construction has relied on computationally intensive bootstrap procedures. We establish a general theorem for the asymptotic distribution of the RESI using a Taylor expansion that accommodates a broad class of models. Simulations under various linear and logistic regression settings show that RESI and its CI have smaller bias and more reliable coverage than commonly used effect sizes such as Cohen's d and f. Combining with robust covariance estimation yields valid inference under model misspecification. We use the methods to investigate associations of depression and behavioral problems with sex and diagnosis in Autism spectrum disorders, and demonstrate that the asymptotic approach achieves up to a 50-fold speedup over the bootstrap. Our work provides a scalable and reliable alternative to bootstrap inference, greatly enhancing the applicability of RESI to high-dimensional studies.

stat.ME↗

arXiv2025

Quality assurance of the Federal Interagency Traumatic Brain Injury Research (FITBIR) database for multi-site MRI analysis

The Federal Interagency Traumatic Brain Injury Research (FITBIR) database is a centralized data repository for traumatic brain injury (TBI) research. It includes over 45,529 magnetic resonance images (MRI) from 6,211 subjects (9,229 imaging sessions) across 26 studies with heterogeneous organization formats, contrasts, acquisition parameters, and demographics. In this work, we organized and harmonized all available structural and diffusion MRI from FITBIR along with relevant demographic information into the Brain Imaging Data Structure. We analyzed whole-brain mean fractional anisotropy, mean diffusivity, total intracranial volume, and the volumes of 132 regions of interest using UNesT segmentations. There were 4,868 subjects (7,035 sessions) with structural MRI and 2,666 subjects (3,763 sessions) with diffusion MRI following quality assurance and harmonization. We modeled profiles for these metrics across ages with generalized additive models for location, scale, and shape (GAMLSS) and found significant differences in subjects with TBI compared to controls in volumes of 15 regions of the brain (q < 0.05, likelihood ratio test with false discovery rate correction).

eess.IV↗

arXiv2023

Analysing spatial point patterns in digital pathology: immune cells in high-grade serous ovarian carcinomas

Multiplex immunofluorescence (mIF) imaging technology facilitates the study of the tumour microenvironment in cancer patients. Due to the capabilities of this emerging bioimaging technique, it is possible to statistically analyse, for example, the co-varying location and functions of multiple different types of immune cells. Complex spatial relationships between different immune cells have been shown to correlate with patient outcomes and may reveal new pathways for targeted immunotherapy treatments. This tutorial reviews methods and procedures relating to spatial point patterns for complex data analysis. We consider tissue cells as a realisation of a spatial point process for each patient. We focus on proper functional descriptors for each observation and techniques that allow us to obtain information about inter-patient variation. Ovarian cancer is the deadliest gynaecological malignancy and can resist chemotherapy treatment effective in cancers. We use a dataset of high-grade serous ovarian cancer samples from 51 patients. We examine the immune cell composition (T cells, B cells, macrophages) within tumours and additional information such as cell classification (tumour or stroma) and other patient clinical characteristics. Our analyses, supported by reproducible software, apply to other digital pathology datasets.

stat.AP↗

arXiv2023

RESI: An R Package for Robust Effect Sizes

Effect size indices are useful parameters that quantify the strength of association and are unaffected by sample size. There are many available effect size parameters and estimators, but it is difficult to compare effect sizes across studies as most are defined for a specific type of population parameter. We recently introduced a new Robust Effect Size Index (RESI) and confidence interval, which is advantageous because it is not model-specific. Here we present the RESI R package, which makes it easy to report the RESI and its confidence interval for many different model classes, with a consistent interpretation across parameters and model types. The package produces coefficient, ANOVA tables, and overall Wald tests for model inputs, appending the RESI estimate and confidence interval to each. The package also includes functions for visualization and conversions to and from other effect size measures. For illustration, we analyze and interpret three different model types.

stat.CO↗

arXiv2021

Accurate confidence interval estimation for non-centrality parameters and effect size indices

We recently proposed a robust effect size index (RESI) that is related to the non-centrality parameter of a test statistic. RESI is advantageous over common indices because (1) it is widely applicable to many types of data; (2) it can rely on a robust covariance estimate; (3) it can accommodate the existence of nuisance parameters. We provided a consistent estimator for the RESI, however, there is no established confidence interval (CI) estimation procedure for the RESI. Here, we use statistical theory and simulations to evaluate several CI estimation procedures for three estimators of the RESI. Our findings show (1) in contrast to common effect sizes, the robust estimator is consistent for the true effect size; (2) common CI procedures for effect sizes that are non-centrality parameters fail to cover the true effect size at the nominal level. Using the robust estimator along with the proposed bootstrap CI is generally accurate and applicable to conduct consistent estimation and valid inference for the RESI, especially when model assumptions may be violated. Based on the RESI, we propose a general framework for the analysis of effect size (ANOES), such that effect sizes and confidence intervals can be easily reported in an analysis of variance (ANOVA) table format for a wide range of models

stat.ME↗

arXiv2021

Faster estimation for constrained gamma mixture models using closed-form estimators

Mixture models are useful in a wide array of applications to identify subpopulations in noisy overlapping distributions. For example, in multiplexed immunofluorescence (mIF), cell image intensities represent expression levels and the cell populations are a noisy mixture of expressed and unexpressed cells. Among mixture models, the gamma mixture model has the strength of being flexible in fitting skewed strictly positive data that occur in many biological measurements. However, the current estimation method uses numerical optimization within the expectation maximization algorithm and is computationally expensive. This makes it infeasible to be applied across many large data sets, as is necessary in mIF data. Powered by a recently developed closed-form estimator for the gamma distribution, we propose a closed-form gamma mixture model that is not only more computationally efficient, but can also incorporate constraints from known biological information to the fitted distribution. We derive the closed-form estimators for the gamma mixture model and use simulations to demonstrate that our model produces comparable results with the current model with significantly less time, and is excellent in constrained model fitting.

stat.ME↗

arXiv2019

A Robust Effect Size Index

Effect size indices are useful tools in study design and reporting because they are unitless measures of association strength that do not depend on sample size. Existing effect size indices are developed for particular parametric models or population parameters. Here, we propose a robust effect size index based on M-estimators. This approach yields an index that is very generalizable because it is unitless across a wide range of models. We demonstrate that the new index is a function of Cohen's $d$, $R^2$, and standardized log odds ratio when each of the parametric models is correctly specified. We show that existing effect size estimators are biased when the parametric models are incorrect (e.g. under unknown heteroskedasticity). We provide simple formulas to compute power and sample size and use simulations to assess the bias and variance of the effect size estimator in finite samples. Because the new index is invariant across models, it has the potential to make communication and comprehension of effect size uniform across the behavioral sciences.

stat.ME↗