The Importance of Model Inspection for Better Understanding Performance Characteristics of Graph Neural Networks

TL;DR

This paper emphasizes the importance of detailed model inspection in understanding the performance and biases of graph neural networks, especially in medical imaging tasks like brain shape classification.

Contribution

It introduces a model inspection framework that reveals how modelling choices affect feature learning and biases in graph neural networks.

Findings

Significant differences in feature embeddings across layers.

Test accuracy alone does not reveal model biases.

Mesh registration impacts feature learning and model performance.

Abstract

This study highlights the importance of conducting comprehensive model inspection as part of comparative performance analyses. Here, we investigate the effect of modelling choices on the feature learning characteristics of graph neural networks applied to a brain shape classification task. Specifically, we analyse the effect of using parameter-efficient, shared graph convolutional submodels compared to structure-specific, non-shared submodels. Further, we assess the effect of mesh registration as part of the data harmonisation pipeline. We find substantial differences in the feature embeddings at different layers of the models. Our results highlight that test accuracy alone is insufficient to identify important model characteristics such as encoded biases related to data source or potentially non-discriminative features learned in submodels. Our model inspection framework offers a valuable tool for practitioners to better understand performance characteristics of deep learning models in medical imaging.