(Sunday, 7/29/2018) 4:30 PM - 5:00 PM
Room: Exhibit Hall | Forum 6
Purpose: The aim of this study performed a study to evaluate head and neck 7T MR unit, with the value of this ultra-high field strength relative to 3T as a standard reference and to compare the localization of clinical target volumes of images of different magnetic field strength.
Methods: All MR measurement was acquired using a human 3T MRI scanner (3.0T; Philips Medical Systems, Netherlands). Additionally, for 7T MR, measurement was produced using a human 7T MRI scanner (7.0T; Philips Medical Systems, Netherlands). Measurements of the images were made using MATLAB Image Processing Toolbox (MATLAB, The MathWorks, Inc., Natick, MA, USA). Both 3T and 7T MRI were used to analyze the image quality, and quantify the loss of information during the distortion process include the widely used objective fidelity criteria which based on mathematically defined assessment approaches, including mean-squared error (MSE), peak signal-to-noise ratio (PSNR), mean absolute error (MAE), and similarity index metric (SSIM) characteristics.
Results: As we acquired ROI as templates to automatically delineate contours on 3T and 7T MR images and placed 7T as the reference, for 3T image, the PSNR was 17.7% greater than that for the 7T. The SSIM is more accurate and consistent than MSE and PSNR. The SSIM of 3T was 36.7% lower than for the 7T image. 7T MRI shown some limitations such as distortion caused by inhomogeneous transmit field treatment accuracy.
Conclusion: In conclusion, MR provides adequate structure information that the treatment planning is most readily achieved such as the head. However, image distortion arises from both system-related effects and object-induced effect and it affects the accuracy of dose calculation. The study clears that before 7T MRI can be used for treatment planning, any image distortion must be quantified and corrected for the treatment accuracy.
Keywords
Image Analysis, Magnetic Fields, Treatment Verification
Taxonomy
IM- Multi-modality imaging systems: CT/MR - human
Contact Email
