Graph Neural Network Learning on the Pediatric Structural Connectome.

Journal: Tomography (Ann Arbor, Mich.)
Published Date:

Abstract

PURPOSE: Sex classification is a major benchmark of previous work in learning on the structural connectome, a naturally occurring brain graph that has proven useful for studying cognitive function and impairment. While graph neural networks (GNNs), specifically graph convolutional networks (GCNs), have gained popularity lately for their effectiveness in learning on graph data, achieving strong performance in adult sex classification tasks, their application to pediatric populations remains unexplored. We seek to characterize the capacity for GNN models to learn connectomic patterns on pediatric data through an exploration of training techniques and architectural design choices.

Authors

  • Anand Srinivasan
    Departments of Radiology, St. Jude Children's Research Hospital, Memphis, TN 38105, USA.
  • Rajikha Raja
    Departments of Radiology, St. Jude Children's Research Hospital, Memphis, TN 38105, USA.
  • John O Glass
    Departments of Radiology, St. Jude Children's Research Hospital, Memphis, TN 38105, USA.
  • Melissa M Hudson
    Department of Oncology, St. Jude Children's Research Hospital, Memphis, TN, United States.
  • Noah D Sabin
    Departments of Radiology, St. Jude Children's Research Hospital, Memphis, TN 38105, USA.
  • Kevin R Krull
    Department of Epidemiology and Cancer Control, St. Jude Children's Research Hospital, Memphis, TN, United States.
  • Wilburn E Reddick
    Departments of Radiology, St. Jude Children's Research Hospital, Memphis, TN 38105, USA.