Does size matter? The relationship between predictive power of single-subject morphometric networks to spatial scale and edge weight

Abstract

Network-level analysis based on anatomical covariance (cortical thickness) has been gaining increasing popularity recently. However, there has not been a systematic study of the impact of spatial scale and edge definitions on predictive performance. In order to obtain a clear understanding of relative performance, there is a need for systematic comparison. In this study, we present a histogram-based approach to construct subjectwise weighted networks that enable a principled comparison across different methods of network analysis. We design several weighted networks based on two large publicly available datasets and perform a robust evaluation of their predictive power under three levels of separability. One of the interesting insights include the robust predictive power resulting from lack of significant impact of changes in nodal size (spatial scale) among the three classification experiments. We also release an open source python package to enable others to implement presented network feature extraction algorithm in their research.

graynet: single-subject morphometric networks for neuroscience connectivity applications

“…When we individualize this approach to enable extraction of single-subject network features, they further enriched insights into abnormalities due to disease (Tijms, Seriès, Willshaw, & Lawrie, 2012,Raamana et al (2015, Palaniyappan, Park, Balain, Dangi, & Liddle (2015), Xu et al (2017)). Moreover, these network-level features demonstrated potential for prognostic applications (Raamana et al, 2015,Raamana et al (2014), in addition to being robust to changes in scale and edge weight metrics (Raamana & Strother, 2017a).…”

Histogram-weighted Networks for Feature Extraction, Connectivity and Advanced Analysis in Neuroscience

“…Network-level analysis of various features, especially if it can be individualized for a singlesubject, is proving to be a valuable tool in many applications (Raamana and Strother 2017;Evans 2013;Palaniyappan et al 2015;Tijms et al 2012;Xu et al 2017;Raamana et al 2015;Lerch et al 2006;He, Chen, and Evans 2007). This package extracts single-subject (individualized, or intrinsic) networks from node-wise data by computing the edge weights based on histogram distance between the distributions of values within each node.…”