Depthgram: Visualizing outliers in high‐dimensional functional data with application to fMRI data exploration

Abstract

Functional magnetic resonance imaging (fMRI) is a non‐invasive technique that facilitates the study of brain activity by measuring changes in blood flow. Brain activity signals can be recorded during the alternate performance of given tasks, that is, task fMRI (tfMRI), or during resting‐state, that is, resting‐state fMRI (rsfMRI), as a measure of baseline brain activity. This contributes to the understanding of how the human brain is organized in functionally distinct subdivisions. fMRI experiments from high‐resolution scans provide hundred of thousands of longitudinal signals for each individual, corresponding to brain activity measurements over each voxel of the brain along the duration of the experiment. In this context, we propose novel visualization techniques for high‐dimensional functional data relying on depth‐based notions that enable computationally efficient 2‐dim representations of fMRI data, which elucidate sample composition, outlier presence, and individual variability. We believe that this previous step is crucial to any inferential approach willing to identify neuroscientific patterns across individuals, tasks, and brain regions. We present the proposed technique via an extensive simulation study, and demonstrate its application on a motor and language tfMRI experiment.

Novel whitening approaches in functional settings

“…However, losing the correlatedness property of functional data over time carries some consequences: if one could assume Gaussianity in the original space, then no interesting projections in a non‐Gaussian sense would be found and rather one would obtain white noise with no practical use. The prior application of robust techniques to detect and remove unusual functional observations with influential behavior (see, for example, Alemán‐Gómez et al, 2022; Arribas‐Gil & Romo, 2014; Navarro‐Esteban & Cuesta‐Albertos, 2021) or a functional principal component data reduction, can eventually enhance the suitability of some statistical techniques based on a whitening transformation.…”

Brain network connectivity underlying remission in early psychosis: a whole brain model approach

“…Image quality assessment, including visual inspection and quality control metrics, ensured dataset quality. More details can be found in a previous work 32 and in supplementary materials.…”

Robust multivariate quality control charts for enhanced variability monitoring

“…Proposed 3 : depending on the family of “concentration algorithms” (CA) by Olive and Hawkins (OH), and using the proposed techniques for (CA) by Olive and Hawkins, a fast, consistent, and highly outlier‐resistant approach, 34 the points drawn on the chart are calculated for the proposed chart is first created (phase I). To estimate the variance‐covariance matrix, trial estimates are first produced, and then attractors are constructed using the (CA) technique from each trial fit.…”