Real-time 3D motion estimation from undersampled MRI using multi-resolution neural networks

Purpose: To enable real-time adaptive magnetic resonance imaging-guided radiotherapy (MRIgRT) by obtaining time-resolved three-dimensional (3D) deformation vector fields (DVFs) with high spatiotemporal resolution and low latency (<500 ms). Theory and Methods: Respiratory-resolved T1-weighted 4D-MRI of 27 patients with lung cancer were acquired using a golden-angle radial stack-of-stars readout. A multiresolution convolutional neural network (CNN) called TEMPEST was trained on up to 32x retrospectively undersampled MRI of 17 patients, reconstructed with a nonuniform fast Fourier transform, to learn optical flow DVFs. TEMPEST was validated using 4D respiratory-resolved MRI, a digital phantom, and a physical motion phantom. The time-resolved motion estimation was evaluated in-vivo using two volunteer scans, acquired on a hybrid MR-scanner with integrated linear accelerator. Finally, we evaluated the model robustness on a publicly-available four-dimensional computed tomography (4D-CT) dataset. Results: TEMPEST produced accurate DVFs on respiratory-resolved MRI at 20-fold acceleration, with the average end-point-error <2 mm, both on respiratory-sorted MRI and on a digital phantom. TEMPEST estimated accurate time-resolved DVFs on MRI of a motion phantom, with an error <2 mm at 28x undersampling. On two volunteer scans, TEMPEST accurately estimated motion compared to the self-navigation signal using 50 spokes per dynamic (366x undersampling). At this undersampling factor, DVFs were estimated within 200 ms, including MRI acquisition. On fully sampled CT data, we achieved a target registration error of 1.87 +/- 1.65 mm without retraining the model. Conclusion: A CNN trained on undersampled MRI produced accurate 3D DVFs with high spatiotemporal resolution for MRIgRT.

Automated summary

This study aims to enable real-time adaptive magnetic resonance imaging-guided radiotherapy (MRIgRT) by obtaining accurate 3D deformation vector fields (DVFs) with high spatiotemporal resolution and low latency. A multiresolution convolutional neural network (CNN) trained on respiratory-resolved T1-weighted 4D-MRI data achieved accurate motion estimation and spatial acceleration. The model demonstrated robustness and accuracy on digital and physical phantoms, as well as in-vivo scans on volunteers, showcasing its potential for clinical applications.