Technical development and In Silico implementation of SyntheticMR in head and neck adaptive radiation therapy: A...

Area of Science:

• Medical Imaging
• Radiation Oncology
• Quantitative MRI

Background:

• SyntheticMR generates quantitative relaxometry maps and synthetic contrast-weighted MRI images rapidly.
• Its feasibility on MR-Simulation and MR-Linac systems for head and neck adaptive radiation oncology workflows is unexplored.

Purpose of the Study:

• To demonstrate the feasibility of SyntheticMR on MR-Simulation and MR-Linac systems.
• To enable rapid quantitative biomarker extraction for adaptive radiation therapy decision-making.

Main Methods:

• Scanned phantoms, volunteers, and a patient using SyntheticMR on MR-Simulation and MR-Linac (4-6 min scans).
• Determined correlation between measured and reference quantitative T1, T2, and PD values.
• Evaluated SyntheticMR for differentiating tumor and normal tissue using organ-at-risk contours and cluster analysis.

Main Results:

• High correlation (Lin's CCC > 0.97) between measured and reference phantom values for both systems.
• Minimal distortion and low mean bias (<11%) for T1, T2, and PD values.
• Significant cluster separability (p < 0.001) between GTV and tongue using SigClust.

Conclusions:

• Technical feasibility of SyntheticMR confirmed for MR-Simulation and MR-Linac.
• SyntheticMR can enrich quantitative biomarkers in head and neck adaptive radiation therapy.