Influence of Processing Pipeline on Cortical Thickness Measurement

Abstract

In recent years, replicability of neuroscientific findings, specifically those concerning correlates of morphological properties of gray matter (GM), have been subject of major scrutiny. Use of different processing pipelines and differences in their estimates of the macroscale GM may play an important role in this context. To address this issue, here, we investigated the cortical thickness estimates of three widely used pipelines. Based on analyses in two independent large-scale cohorts, we report high levels of within-pipeline reliability of the absolute cortical thickness-estimates and comparable spatial patterns of cortical thickness-estimates across all pipelines. Within each individual, absolute regional thickness differed between pipelines, indicating that in-vivo thickness measurements are only a proxy of actual thickness of the cortex, which shall only be compared within the same software package and thickness estimation technique. However, at group level, cortical thickness-estimates correlated strongly between pipelines, in most brain regions. The smallest between-pipeline correlations were observed in para-limbic areas and insula. These regions also demonstrated the highest interindividual variability and the lowest reliability of cortical thickness-estimates within each pipeline, suggesting that structural variations within these regions should be interpreted with caution.

Keywords: in-vivo cortical thickness; interindividual variability; reliability; replicability; software comparison.

Figure 1

Distribution of global cortical thickness across samples and pipelines. FS: FreeSurfer.

Figure 2

Regional mean (left columns) and SD (right columns) of cortical thickness estimates of each pipeline. For each pipeline, parcels with the highest SD (top 10%) are depicted with a black surrounding.

Figure 3

Regional between-pipeline comparison of cortical thickness estimates within the HCP sample. Top rows: paired-difference map of cortical thickness estimates; red-yellow colors depict higher thickness estimate of the pipeline mentioned in column versus row and the dark-light blue depicts the opposite direction. Only regions with significant paired t-test comparisons are colored (P < 4 × 10⁻⁵); lower rows: between-pipeline regional Spearman’s correlation.

Figure 4

Regional between-pipeline comparison of cortical thickness estimates within the GOBS sample. Top rows: paired-difference map of cortical thickness estimates; red-yellow colors depict higher thickness estimate of the pipeline mentioned in column versus row and the dark-light blue depicts the opposite direction. Only regions with significant paired t-test comparisons are colored (P < 4 × 10⁻⁵); lower rows: between-pipeline regional Spearman’s correlation.

Figure 5

Between-pipeline comparison of global cortical thickness and its interindividual variability. Scatter plots of pair-wise comparison of global cortical estimates and their Spearman’s correlations, within two cohorts. (A, C) Dashed line depicts the identity line(y = x). (B, D) Scatter plots of association between participants’ age and global cortical thickness from each pipeline, for each cohort. Legends depict Spearman’s correlation between age and global thickness of each pipeline.

Figure 6

Test–retest reliability. Left: pairs of violin plots demonstrate distribution of global cortical thickness estimates from odd and even scans, over all the subjects. Right: regional distribution of MAPE for each pipeline. The top ten percent of the parcels, showing the lowest test–retest reliability, are surrounded with black line. The lighter colors in the spatial maps depict higher MAPE.