Real-Time 2D MR Cine From Beam Eye’s View With Tumor-Volume Projection to Ensure Beam-to-Tumor Conformality for MR-Guided Radiotherapy of Lung Cancer

Abstract

PurposeTo minimize computation latency using a predictive strategy to retrieve and project tumor volume onto 2D MR beam eye’s view (BEV) cine from time-resolved four-dimensional magnetic resonance imaging (TR-4DMRI) libraries (inhalation/exhalation) for personalized MR-guided intensity-modulated radiotherapy (IMRT) or volumetric-modulated arc therapy (VMAT).MethodsTwo time-series forecasting algorithms, autoregressive (AR) modeling and deep-learning-based long short-term memory (LSTM), were applied to predict the diaphragm position in the next 2D BEV cine to identify a motion-matched and hysteresis-accounted image to retrieve the tumor volume from the inhalation/exhalation TR-4DMRI libraries. Three 40-s TR-4DMRI (2 Hz, 3 × 80 images) per patient of eight lung cancer patients were used to create patient-specific inhalation/exhalation 4DMRI libraries, extract diaphragmatic waveforms, and interpolate them to f = 4 and 8 Hz to match 2D cine frame rates. Along a (40•f)-timepoint waveform, 30•f training timepoints were moved forward to produce 3×(10•f-1) predictions. The accuracy of position prediction was assessed against the waveform ground truth. The accuracy of tumor volume projections was evaluated using the center-of-mass difference (∆COM) and Dice similarity index against the TR-4DMRI ground truth for both IMRT (six beam angles, 30° interval) and VMAT (240/480 beam angles, 1.5°/0.75° interval, at 4/8 Hz, respectively).ResultsThe accuracy of the first-timepoint prediction is 0.36 ± 0.10 mm (AR) and 0.62 ± 0.21 mm (LSTM) at 4 Hz and 0.06 ± 0.02 mm (AR) and 0.18 ± 0.06 mm (LSTM) at 8 Hz. A 10%–20% random error in prediction-library matching increases the overall uncertainty slightly. For both IMRT and VMAT, the accuracy of projected tumor volume contours on 2D BEV cine is ∆COM = 0.39 ± 0.13 mm and DICE = 0.97 ± 0.02 at 4 Hz and ∆COM = 0.10 ± 0.04 mm and DICE = 1.00 ± 0.00 at 8Hz.ConclusionThis study demonstrates the feasibility of accurately predicting respiratory motion during 2D BEV cine imaging, identifying a motion-matched and hysteresis-accounted tumor volume, and projecting tumor volume contour on 2D BEV cine for real-time assessment of beam-to-tumor conformality, promising for optimal personalized MR-guided radiotherapy.

Deep learning in MRI‐guided radiation therapy: A systematic review

“…instead uses autoregression and the LSTM time series modeling to predict the diaphragm position and to find the matching 4D MRI volume. Autoregression outperforms an LSTM model which could be attributed to a low number of patients 152 . Patient motion is alternatively predicted in Terpestra et al.…”

“…Autoregression outperforms an LSTM model which could be attributed to a low number of patients. 152 Patient motion is alternatively predicted in Terpestra et alby using undersampled 3D cine MRI to generate the DVF with a CNN with low target registration error. 153 Similarly, Romaguera et al…”

Real-time 4D MRI using MR signature matching (MRSIGMA) on a 1.5T MR-Linac system

“…This limitation has recently been addressed by utilizing beam-eye-view 2D cine imaging with tumor volume projection adapted to each radiation beam (gantry) angle, enabling intra-fractional monitoring of tumor motion and verification of beam conformality (Nie et al 2021). A subsequent study proposed a predictive strategy to minimize latency in identifying motion-matched tumor volume while accounting for hysteresis, facilitating real-time assessment of beam-to-tumor conformality (Nie and Li 2022). While the beam-eye-view approach shows promising results for MR-guided radiotherapy, its application has only been demonstrated in lung cancer patients using a 3T MRI scanner.…”