Joint radial trajectory correction for accelerated T₂ ^* mapping on an MR-Linac

Area of Science:

• Medical Imaging
• Radiotherapy Physics
• Quantitative MRI

Background:

• T2* mapping can characterize tumor hypoxia, a factor in therapy resistance.
• Acquiring T2* maps during MR-guided radiotherapy could enable treatment adaptation.
• Hypoxia-associated therapy resistance may be overcome by dose escalation to resistant sub-volumes.

Purpose of the Study:

• To demonstrate the feasibility of accelerated T2* mapping using model-based reconstruction with integrated trajectory auto-correction (TrACR).
• To validate the technique for MR-guided radiotherapy on an MR-Linear accelerator (MR-Linac).

Main Methods:

• Validated a novel accelerated T2* mapping technique using model-based reconstruction with integrated trajectory auto-correction (TrACR).
• Compared sequential and joint T2* mapping approaches in numerical phantoms with varying noise levels and gradient delays.
• Acquired in vivo data from prostate and head and neck cancer patients undergoing MR-Linac treatment, comparing reconstructed T2* maps with and without trajectory corrections.

Main Results:

• Numerical simulations showed the joint approach with TrACR reduced errors in T2* maps compared to uncorrected and sequential methods across all noise levels.
• In vivo data demonstrated that TrACR resulted in fewer artifacts and improved visual quality of T2* maps.
• Reconstructed T2* maps from different treatment fractions revealed changes within the planning target volume (PTV) for both patient types.

Conclusions:

• The proposed approach enables retrospective, data-driven gradient delay correction, crucial for hybrid MR-Linac devices.
• Accelerated T2* maps were acquired in under 5 minutes, suitable for integration into MR-guided radiotherapy workflows.
• This technique supports online adaptive radiotherapy by minimizing patient burden and allowing time for additional imaging.