Radiation and Cancer Physics

PV 01 - Poster Viewing Q&A - Session 1

SU_35_3596 - Evaluation of the Geometric Distortion in Clinical MR Sequences for a 1.5T MR-Linac System

Sunday, September 15
1:15 PM - 2:30 PM
Location: ASTRO Innovation Hub

Evaluation of the Geometric Distortion in Clinical MR Sequences for a 1.5T MR-Linac System
M. Aima, J. M. Slagowski, Y. Ding, S. Vedam, J. Yang, K. Hwang, J. Szklaruk, S. Choi, C. D. Fuller, C. Chung, and J. Wang; The University of Texas MD Anderson Cancer Center, Houston, TX

Purpose/Objective(s): Geometric distortion can be a considerable source of diagnostic as well as dosimetric uncertainty for an MR-Linac system. The focus of this study was to assess the geometric distortion observed in a 1.5T MR-Linac machine using various clinical MR sequences.

Materials/Methods: A manufacturer provided cylindrical geometric distortion phantom with a marker spacing of 25 mm (medial-lateral) x 25 mm (anterior-posterior) x 55 mm (superior-inferior) was used for this work. The diameter, and the length of the phantom were 500 mm and 330 mm respectively. MR images of the phantom were acquired using a manufacturer-provided MR protocol and ten additional MR protocols including Brain, Head and Neck, Thorax, and Pelvis clinical sequences. A custom geometric distortion software code was developed for the analysis of the phantom images acquired using the clinical sequences. The code was initially validated against the manufacturer provided software for the default non-clinical protocol analysis. Wrap-around artifacts and markers detected with less than half their size were not included in the distortion analysis.

Results: The absolute geometric distortion (98th percentile) measured for the Thorax T1 3D, Thorax T2 3D, Head and Neck T1 3D, Head and Neck T2 3D, Pelvis T1 3D, Pelvis T2 3D, Pelvis T1 3D fast, and Pelvis T2 3D fast clinical sequences was 2.7 mm, 2.8 mm, 3.7 mm, 3.4 mm, 4.0 mm, 1.8 mm, 3.2 mm, and 2.2 mm, respectively over a 500 mm diameter of spherical volume (DSV) around the center of the phantom. For the Brain T1 3D and Brain T2 3D clinical sequences the absolute geometric distortion (98th percentile) measured was 2.0 mm and 2.3 mm respectively over a 400 mm DSV. The measured absolute geometric distortion (98th percentile) for all clinical sequences was within 1.1 mm for a 200 mm DSV. Susceptibility artifacts were observed for some sequences. There was a noticeable drop in marker detection due to lower signal-to-noise ratio and signal non-uniformity around the diagonal edges of the acquired slices of the phantom using all the clinical sequences. The absolute mean geometric distortion was within 1.3 mm for all clinical sequences. An example table presenting the results of the geometric distortion (GD) evaluation for the T2 3D Pelvis clinical sequence is provided below.

Conclusion: An investigation was performed to estimate the geometric distortion in clinical MR sequences for an MR-Linac system. The magnitude of measured geometric distortion varied depending on the clinical MR sequence used for image acquisition. Any MR-Linac radiation treatment related decision-making should take this source of uncertainty into consideration.
Diameter of spherical volume
Parameter 200 mm 300 mm 400 mm 500 mm
Mean GD (mm) 0.3 0.5 0.6 0.7
98th percentile GD (mm) 0.8 0.9 1.4 1.8
Max GD (mm) 1.0 1.2 2.2 2.6
Std Dev GD (mm) 0.2 0.2 0.3 0.4

Author Disclosure: M. Aima: None. J.M. Slagowski: None. S. Vedam: None. J. Szklaruk: None. C.D. Fuller: Consultant; Elekta AB, Nederlandse Organisatie voor Wetenschappelijk Onde. Grant funding; Elekta AB. Honoraria; Elekta AB, Nederlandse Organisatie voor Wetenschappelijk Onde. Research Grant; National Institutes of Health, National Science Foundation, Elekta AB. Travel Expenses; Elekta AB, Nederlandse Organisatie voor Wetenschappelijk Onde. Associate Editor; Radiographics. Committee Member; American College of Radiology. Data Management Task Force Committee Member; MR-LinAc Consortium. Member; National Cancer Institute. Task Group Member; American Association of Physicists in Medicine. J. Wang: Research Grant; Elekta.

Manik Aima, PhD, MS, MSE

MD Anderson Cancer Center

Presentation(s):

Send Email for Manik Aima


Assets

SU_35_3596 - Evaluation of the Geometric Distortion in Clinical MR Sequences for a 1.5T MR-Linac System



Attendees who have favorited this

Please enter your access key

The asset you are trying to access is locked. Please enter your access key to unlock.

Send Email for Evaluation of the Geometric Distortion in Clinical MR Sequences for a 1.5T MR-Linac System