Model-Interpolated Gating for Magnetic Resonance Image-Guided Radiation Therapy

Area of Science:

• Medical Physics
• Radiotherapy Technology
• Medical Imaging

Background:

• Accurate tumor targeting in radiation therapy is crucial for effective treatment and minimizing damage to surrounding tissues.
• Motion during treatment, particularly respiratory motion, poses a significant challenge for precise radiation delivery.
• Current MRI-guided radiation therapy gating methods often require fast imaging, limiting their applicability.

Purpose of the Study:

• To develop and validate a novel technique for radiation therapy gating using slow magnetic resonance imaging (MRI) and a motion model.
• To assess the accuracy and feasibility of this technique for real-time tumor tracking during treatment.
• To explore its potential for T2-weighted imaging-based gating and additional imaging during gated treatments.

Main Methods:

• A physiologically guided breathing motion model was employed to interpolate target positions between slow 2D MRI acquisitions (1-3 seconds apart).
• The model was continuously updated using a respiratory bellows surrogate and the latest 2D images.
• The technique was evaluated using phantom studies and in 8 volunteers on a 0.35T MRI-guided radiation therapy system, with balanced steady-state free precession and T2-weighted sequences.

Main Results:

• Phantom studies demonstrated high gating accuracy (98.3%) and beam-on positive predictive value (PPV) (98.4%), with a median absolute distance of 0.33 mm.
• Volunteer studies showed comparable results, with an average accuracy of 93.7% and PPV of 92.1% (median distance 0.86 mm).
• T2-weighted gating in volunteers achieved an average accuracy of 94.3% and PPV of 92.5%.

Conclusions:

• The developed model-interpolated gating technique is a promising concept for MRI-guided radiation therapy.
• The technique demonstrated sufficient accuracy for potential clinical implementation.
• Further research is needed to address out-of-plane motion and integrate internal MR-based respiratory surrogates.