Deep Learning for Quality Control of Subcortical Brain 3D Shape Models

TL;DR

This paper develops deep learning models using CNN architectures to evaluate the quality of subcortical brain shape models from MRI data, significantly reducing human rater time.

Contribution

It introduces a novel geometry feature augmentation technique and a model decision visualization method for improved quality assessment of brain shape models.

Findings

ResNet outperforms other CNN architectures in accuracy.

Models reduce human rater time by up to 70%.

Effective geometry feature augmentation improves model performance.

Abstract

We present several deep learning models for assessing the morphometric fidelity of deep grey matter region models extracted from brain MRI. We test three different convolutional neural net architectures (VGGNet, ResNet and Inception) over 2D maps of geometric features. Further, we present a novel geometry feature augmentation technique based on a parametric spherical mapping. Finally, we present an approach for model decision visualization, allowing human raters to see the areas of subcortical shapes most likely to be deemed of failing quality by the machine. Our training data is comprised of 5200 subjects from the ENIGMA Schizophrenia MRI cohorts, and our test dataset contains 1500 subjects from the ENIGMA Major Depressive Disorder cohorts. Our final models reduce human rater time by 46-70%. ResNet outperforms VGGNet and Inception for all of our predictive tasks.