PLoS Biology (Apr 2022)

Population heterogeneity in clinical cohorts affects the predictive accuracy of brain imaging.

  • Oualid Benkarim,
  • Casey Paquola,
  • Bo-Yong Park,
  • Valeria Kebets,
  • Seok-Jun Hong,
  • Reinder Vos de Wael,
  • Shaoshi Zhang,
  • B T Thomas Yeo,
  • Michael Eickenberg,
  • Tian Ge,
  • Jean-Baptiste Poline,
  • Boris C Bernhardt,
  • Danilo Bzdok

DOI
https://doi.org/10.1371/journal.pbio.3001627
Journal volume & issue
Vol. 20, no. 4
p. e3001627

Abstract

Read online

Brain imaging research enjoys increasing adoption of supervised machine learning for single-participant disease classification. Yet, the success of these algorithms likely depends on population diversity, including demographic differences and other factors that may be outside of primary scientific interest. Here, we capitalize on propensity scores as a composite confound index to quantify diversity due to major sources of population variation. We delineate the impact of population heterogeneity on the predictive accuracy and pattern stability in 2 separate clinical cohorts: the Autism Brain Imaging Data Exchange (ABIDE, n = 297) and the Healthy Brain Network (HBN, n = 551). Across various analysis scenarios, our results uncover the extent to which cross-validated prediction performances are interlocked with diversity. The instability of extracted brain patterns attributable to diversity is located preferentially in regions part of the default mode network. Collectively, our findings highlight the limitations of prevailing deconfounding practices in mitigating the full consequences of population diversity.