Angular Super-Resolution in Diffusion MRI with a 3D Recurrent Convolutional Autoencoder

TL;DR

This paper introduces a 3D recurrent convolutional autoencoder that significantly improves angular super-resolution in diffusion MRI, enabling higher resolution imaging from limited clinical scan data.

Contribution

The work presents a novel 3D RCNN model with ConvLSTM cells for angular super-resolution in diffusion MRI, outperforming traditional interpolation methods.

Findings

3D RCNN achieves lowest error rates in super-resolution tasks.

Model performs best at very low angular resolutions.

Code is publicly available for reproducibility.

Abstract

High resolution diffusion MRI (dMRI) data is often constrained by limited scanning time in clinical settings, thus restricting the use of downstream analysis techniques that would otherwise be available. In this work we develop a 3D recurrent convolutional neural network (RCNN) capable of super-resolving dMRI volumes in the angular (q-space) domain. Our approach formulates the task of angular super-resolution as a patch-wise regression using a 3D autoencoder conditioned on target b-vectors. Within the network we use a convolutional long short term memory (ConvLSTM) cell to model the relationship between q-space samples. We compare model performance against a baseline spherical harmonic interpolation and a 1D variant of the model architecture. We show that the 3D model has the lowest error rates across different subsampling schemes and b-values. The relative performance of the 3D RCNN is greatest in the very low angular resolution domain. Code for this project is available at https://github.com/m-lyon/dMRI-RCNN.