On Real-time Image Reconstruction with Neural Networks for MRI-guided Radiotherapy

TL;DR

This paper demonstrates that AUTOMAP neural networks can rapidly and accurately reconstruct MRI images from undersampled data, enabling real-time image guidance in MRI-guided radiotherapy, with improved motion robustness.

Contribution

The study introduces a neural network-based approach using AUTOMAP for fast, real-time MRI reconstruction from undersampled data, outperforming traditional methods in speed and robustness.

Findings

AUTOMAP achieves equivalent accuracy to compressed sensing with faster processing times.

Motion-trained models improve target tracking accuracy in dynamic lung tumor simulations.

Neural network reconstruction shows potential for real-time MRI-guided radiotherapy applications.

Abstract

MRI-guidance techniques that dynamically adapt radiation beams to follow tumor motion in real-time will lead to more accurate cancer treatments and reduced collateral healthy tissue damage. The gold-standard for reconstruction of undersampled MR data is compressed sensing (CS) which is computationally slow and limits the rate that images can be available for real-time adaptation. Here, we demonstrate the use of automated transform by manifold approximation (AUTOMAP), a generalized framework that maps raw MR signal to the target image domain, to rapidly reconstruct images from undersampled radial k-space data. The AUTOMAP neural network was trained to reconstruct images from a golden-angle radial acquisition, a benchmark for motion-sensitive imaging, on lung cancer patient data and generic images from ImageNet. Model training was subsequently augmented with motion-encoded k-space data derived from videos in the YouTube-8M dataset to encourage motion robust reconstruction. We find that AUTOMAP-reconstructed radial k-space has equivalent accuracy to CS but with much shorter processing times after initial fine-tuning on retrospectively acquired lung cancer patient data. Validation of motion-trained models with a virtual dynamic lung tumor phantom showed that the generalized motion properties learned from YouTube lead to improved target tracking accuracy. Our work shows that AUTOMAP can achieve real-time, accurate reconstruction of radial data. These findings imply that neural-network-based reconstruction is potentially superior to existing approaches for real-time image guidance applications.