Room: Karl Dean Ballroom B1
Purpose: MRI-only radiotherapy treatment planning is attractive since MRI provides superior soft tissue contrast without ionizing radiation compared with CT. However, it requires the generation of synthetic CT (SCT) from MRI for patient setup and dose calculation. Our method based on machine learning and auto-context model has been demonstrated to provide SCT with excellent image quality, while its dose calculation accuracy remains an open question. In this study, we aim to investigate the accuracy of dose calculation in brain frameless stereotactic radiosurgery (SRS) using SCT which are generated from MRI using the learning-based method.
Methods: We retrospectively investigated a total of 19 treatment plans from 14 patients, each of whom has CT and MRI acquired before treatment. The SCT was registered to planning CT for generating SCT-based treatment plans. The original planning CT-based plans served as ground truth. Clinically-relevant dose volume histogram (DVH) metrics were extracted from both ground truth and SCT-based plans for comparison and evaluation. Gamma analysis was performed for the comparison of absorbed dose distributions between the both plans of each patient.
Results: The side-by-side comparisons on image quality and dose distributions demonstrated very good agreement of image contrast and calculated dose between SCT and CT. The average differences in DVH metrics for planning target volumes (PTVs) were less than 0.6%, and no differences in those for organs-at-risk (OAR) at a significance level of 0.05. The average pass rate of gamma analysis was over 99%.
Conclusion: These quantitative results indicate that the SCT generated from MRI using our proposed learning-based method are accurate enough to replace standard CT simulation images for dose calculation in brain SRS treatment. This study also demonstrates the great potential for MRI to completely replace CT scans in the process of simulation and treatment planning.
Not Applicable / None Entered.