Outlier Detection for Multi-Network Data

TL;DR

This paper introduces ODIN, a new method for detecting outliers in multi-network neuroimaging data, which improves the reliability of subsequent analyses by identifying poor-quality or bizarre brain network matrices.

Contribution

The paper presents ODIN, a hierarchical generalized linear model-based outlier detection method specifically designed for multi-network data in neuroscience, addressing a previously overlooked problem.

Findings

ODIN effectively identifies moderate to extreme outliers in simulated and real data.

Removing outliers with ODIN significantly alters downstream analysis results.

The method demonstrates computational efficiency and practical utility in neuroimaging studies.

Abstract

It has become routine in neuroscience studies to measure brain networks for different individuals using neuroimaging. These networks are typically expressed as adjacency matrices, with each cell containing a summary of connectivity between a pair of brain regions. There is an emerging statistical literature describing methods for the analysis of such multi-network data in which nodes are common across networks but the edges vary. However, there has been essentially no consideration of the important problem of outlier detection. In particular, for certain subjects, the neuroimaging data are so poor quality that the network cannot be reliably reconstructed. For such subjects, the resulting adjacency matrix may be mostly zero or exhibit a bizarre pattern not consistent with a functioning brain. These outlying networks may serve as influential points, contaminating subsequent statistical analyses. We propose a simple Outlier DetectIon for Networks (ODIN) method relying on an influence measure under a hierarchical generalized linear model for the adjacency matrices. An efficient computational algorithm is described, and ODIN is illustrated through simulations and an application to data from the UK Biobank. ODIN was successful in identifying moderate to extreme outliers. Removing such outliers can significantly change inferences in downstream applications.