A Comparative Study of Graph Neural Networks for Shape Classification in Neuroimaging
This work provides guidance for practitioners in neuroimaging on optimizing GNNs for shape-based tasks like disease detection, but it is incremental as it compares existing methods.
The study tackled the problem of unclear architectural choices for graph neural networks in medical shape classification by conducting a comparative analysis, finding that FPFH node features improve performance and generalization, with results validated on Alzheimer's disease classification.
Graph neural networks have emerged as a promising approach for the analysis of non-Euclidean data such as meshes. In medical imaging, mesh-like data plays an important role for modelling anatomical structures, and shape classification can be used in computer aided diagnosis and disease detection. However, with a plethora of options, the best architectural choices for medical shape analysis using GNNs remain unclear. We conduct a comparative analysis to provide practitioners with an overview of the current state-of-the-art in geometric deep learning for shape classification in neuroimaging. Using biological sex classification as a proof-of-concept task, we find that using FPFH as node features substantially improves GNN performance and generalisation to out-of-distribution data; we compare the performance of three alternative convolutional layers; and we reinforce the importance of data augmentation for graph based learning. We then confirm these results hold for a clinically relevant task, using the classification of Alzheimer's disease.