Combining imaging and shape features for prediction tasks of Alzheimer's disease classification and brain age regression

TL;DR

This paper presents a model that combines MRI image features and brain shape information using neural networks to improve predictions of brain age and Alzheimer's disease classification, showing significant performance gains.

Contribution

It introduces a novel fusion of ResNet image embeddings with shape embeddings from a graph neural network for enhanced brain analysis tasks.

Findings

Improved prediction accuracy for brain age and Alzheimer's classification.

Shape features contribute significantly to model performance.

Effective fusion of imaging and shape data demonstrated on multiple datasets.

Abstract

We investigate combining imaging and shape features extracted from MRI for the clinically relevant tasks of brain age prediction and Alzheimer's disease classification. Our proposed model fuses ResNet-extracted image embeddings with shape embeddings from a bespoke graph neural network. The shape embeddings are derived from surface meshes of 15 brain structures, capturing detailed geometric information. Combined with the appearance features from T1-weighted images, we observe improvements in the prediction performance on both tasks, with substantial gains for classification. We evaluate the model using public datasets, including CamCAN, IXI, and OASIS3, demonstrating the effectiveness of fusing imaging and shape features for brain analysis.