Improved Brain Age Estimation with Slice-based Set Networks

TL;DR

This paper introduces a novel slice-based set network architecture for brain age estimation from MRI scans, which improves training speed and prediction accuracy over existing methods by encoding slices with 2D CNNs and aggregating with permutation invariant layers.

Contribution

The paper proposes a new architecture combining 2D slice encoding with set networks for brain age prediction, reducing training time and enhancing accuracy.

Findings

Faster training with permutation invariant layers

Improved prediction accuracy over state-of-the-art methods

Effective use of 2D slices for 3D MRI analysis

Abstract

Deep Learning for neuroimaging data is a promising but challenging direction. The high dimensionality of 3D MRI scans makes this endeavor compute and data-intensive. Most conventional 3D neuroimaging methods use 3D-CNN-based architectures with a large number of parameters and require more time and data to train. Recently, 2D-slice-based models have received increasing attention as they have fewer parameters and may require fewer samples to achieve comparable performance. In this paper, we propose a new architecture for BrainAGE prediction. The proposed architecture works by encoding each 2D slice in an MRI with a deep 2D-CNN model. Next, it combines the information from these 2D-slice encodings using set networks or permutation invariant layers. Experiments on the BrainAGE prediction problem, using the UK Biobank dataset, showed that the model with the permutation invariant layers trains faster and provides better predictions compared to other state-of-the-art approaches.