Conditional-Based Transformer Network With Learnable Queries for 4D Deformation Forecasting and Tracking.

Real-time motion management for image-guided radiation therapy interventions plays an important role for accurate dose delivery. Forecasting future 4D deformations from in-plane image acquisitions is fundamental for accurate dose delivery and tumor targeting. However, anticipating visual representations is challenging and is not exempt from hurdles such as the prediction from limited dynamics, and the high-dimensionality inherent to complex deformations. Also, existing 3D tracking approaches typically need both template and search volumes as inputs, which are not available during real-time treatments. In this work, we propose an attention-based temporal prediction network where features extracted from input images are treated as tokens for the predictive task. Moreover, we employ a set of learnable queries, conditioned on prior knowledge, to predict future latent representation of deformations. Specifically, the conditioning scheme is based on estimated time-wise prior distributions computed from future images available during the training stage. Finally, we propose a new framework to address the problem of temporal 3D local tracking using cine 2D images as inputs, by employing latent vectors as gating variables to refine the motion fields over the tracked region. The tracker module is anchored on a 4D motion model, which provides both the latent vectors and the volumetric motion estimates to be refined. Our approach avoids auto-regression and leverages spatial transformations to generate the forecasted images. The tracking module reduces the error by 63% compared to a conditional-based transformer 4D motion model, yielding a mean error of 1.5± 1.1 mm. Furthermore, for the studied cohort of abdominal 4D MRI images, the proposed method is able to predict future deformations with a mean geometrical error of 1.2± 0.7 mm.

Impact of temporal resolution and motion correction for dynamic contrast-enhanced MRI of the liver using an accelerated golden-angle radial sequence.

This paper presents a prospective study evaluating the impact on image quality and quantitative dynamic contrast-enhanced (DCE)-MRI perfusion parameters when varying the number of respiratory motion states when using an eXtra-Dimensional Golden-Angle Radial Sparse Parallel (XD-GRASP) MRI sequence. DCE acquisition was performed using a 3D stack-of-stars gradient-echo golden-angle radial acquisition in free-breathing with 100 spokes per motion state and temporal resolution of 6 s/volume, and using a non-rigid motion compensation to align different motion states. Parametric analysis was conducted using a dual-input single-compartment model. Nonparametric analysis was performed on the time-intensity curves. A total of 22 hepatocellular carcinomas (size: 11-52 mm) were evaluated. XD-GRASP reconstructed with increasing number of spokes for each motion state increased the signal-to-noise ratio (SNR) (p < 0.05) but decreased temporal resolution (0.04 volume/s vs 0.17 volume/s for one motion state) (p < 0.05). A visual scoring by an experienced radiologist show no change between increasing number of motion states with same number of spokes using the Likert score. The normalized maximum intensity time ratio, peak enhancement ratio and tumor arterial fraction increased with decreasing number of motion states (p < 0.05) while the transfer constant from the portal venous plasma to the surrounding tissue significantly decreased (p < 0.05). These same perfusion parameters show a significant difference in case of tumor displacement more than 1 cm (p < 0.05) whereas in the opposite case there was no significant variation. While a higher number of motion states and higher number of spokes improves SNR, the resulting lower temporal resolution can influence quantitative parameters that capture rapid signal changes. Finally, fewer displacement compensation is advantageous with lower number of motion state due to the higher temporal resolution. XD-GRASP can be used to perform quantitative perfusion measures in the liver, but the number of motion states may significantly alter some quantitative parameters.

Automatic self-gated 4D-MRI construction from free-breathing 2D acquisitions applied on liver images.

PURPOSE: MRI slice reordering is a necessary step when three-dimensional (3D) motion of an anatomical region of interest has to be extracted from multiple two-dimensional (2D) dynamic acquisition planes, e.g., for the construction of motion models used for image-guided radiotherapy. Existing reordering methods focus on obtaining a spatially coherent reconstructed volume for each time. However, little attention has been paid to the temporal coherence of the reconstructed volumes, which is of primary importance for accurate 3D motion extraction. This paper proposes a fully automatic self-sorting four-dimensional MR volume construction method that ensures the temporal coherence of the results. METHODS: First, a pseudo-navigator signal is extracted for each 2D dynamic slice acquisition series. Then, a weighted graph is created using both spatial and motion information provided by the pseudo-navigator. The volume at a given time point is reconstructed following the shortest paths in the graph starting that time point of a reference slice chosen based on its pseudo-navigator signal. RESULTS: The proposed method is evaluated against two state-of-the-art slice reordering algorithms on a prospective dataset of 12 volunteers using both spatial and temporal quality metrics. The automated end-exhale extraction showed results closed to the median value of the manual operators. Furthermore, the results of the validation metrics show that the proposed method outperforms state-of-the-art methods in terms of both spatial and temporal quality. CONCLUSION: Our approach is able to automatically detect the end-exhale phases within one given anatomical position and cope with irregular breathing.