Robust outlier detection for heterogeneous distributions applicable to censoring in functional MRI

2025-05-17

AI Summary

This study addresses the challenge of detecting high-intensity “burst” outliers—high-dimensional, non-Gaussian artifacts induced by head motion—in fMRI data. Methodologically, it introduces, for the first time, the sinh-arcsinh (SHASH) distribution family to construct a normalizing transformation capable of jointly correcting skewness and heavy- or light-tailedness; this is integrated with an improved isolation forest (iForest) for robust anomaly label initialization. Subsequently, robust z-score and minimum covariance determinant (MCD) distance are fused to enhance discriminative reliability. Evaluated on both synthetic data and multi-site real fMRI datasets, the framework demonstrates significant superiority over state-of-the-art baselines under skewed, heavy-tailed, and highly contaminated scenarios. It achieves higher true positive detection rates while maintaining tighter control over false positives, exhibiting both superior performance and stability across diverse outlier regimes.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Adversarial Learning & Robustness
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Functional magnetic resonance imaging (fMRI) data are prone to intense"burst"noise artifacts due to head movements and other sources. Such volumes can be considered as high-dimensional outliers that can be identified using statistical outlier detection techniques, which allows for controlling the false positive rate. Previous work has used dimension reduction and multivariate outlier detection techniques, including the use of robust minimum covariance determinant (MCD) distances. Under Gaussianity, the distribution of these robust distances can be approximated, and an upper quantile of that distribution can be used to identify outlying volumes. Unfortunately, the Gaussian assumption is unrealistic for fMRI data in this context. One way to address this is to transform the data to Normality. A limitation of existing robust methods for this purpose, such as robust Box-Cox and Yeo-Johnson transformations, is that they can deal with skew but not heavy or light tails. Here, we develop a novel robust method for transformation to central Normality based on the highly flexible sinh-arcsinh (SHASH) family of distributions. To avoid the influence of outliers, it is crucial to initialize the outlier labels with a high degree of sensitivity. For this purpose, we consider a commonplace robust z-score approach, and a modified isolation forest (iForest) approach, a popular technique for anomaly detection in machine learning. Through extensive simulation studies, we find that our proposed SHASH transformation initialized using iForest clearly outperforms benchmark methods in a variety of settings, including skewed and heavy tailed distributions, and light to heavy outlier contamination. We also apply the proposed techniques to several example datasets and find this combination to have consistently strong performance.

Problem

• Detects outliers in fMRI data with burst noise artifacts
• Addresses non-Gaussian distributions using SHASH transformation
• Improves robustness with modified isolation forest initialization

Innovation

• Uses SHASH family for robust normality transformation
• Initializes outlier labels with modified iForest
• Combines SHASH and iForest for superior performance