First Take a Picture First Before Uploading Phantom
J Appl Clin Med Phys. 2021 November; 22(eleven): 21–28.
First clinical feel of correcting phantom‐based prototype distortion related to gantry position on a 0.35T MR‐Linac
Benjamin C. Lewis
1 Section of Radiation Oncology, Washington University School of Medicine, St. Louis, Missouri USA,
Jaeik Shin
1 Section of Radiation Oncology, Washington University Schoolhouse of Medicine, St. Louis, Missouri USA,
2 Department of Radiations Oncology, Yonsei University College of Medicine, Seoul Commonwealth of Korea,
Benjamin Quinn
3 Department for the Modus Medical, Modus Medical Devices Inc., London Ontario, Canada,
Enzo Barberi
3 Department for the Modus Medical, Modus Medical Devices Inc., London Ontario, Canada,
Domenic Sievert
ane Section of Radiation Oncology, Washington University Schoolhouse of Medicine, St. Louis, Missouri USA,
Jin Sung Kim
2 Department of Radiation Oncology, Yonsei University Higher of Medicine, Seoul Commonwealth of Korea,
Taeho Kim
1 Department of Radiation Oncology, Washington University School of Medicine, St. Louis, Missouri USA,
Received 2021 April 2; Revised 2021 Jun 17; Accustomed 2021 Aug vii.
- Data Availability Statement
-
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Abstract
MR‐guided radiotherapy requires strong imaging spatial integrity to deliver high quality plans and provide accurate dose calculation. The MRI system, nevertheless, can exist compromised by the integrated linear accelerator (Linac), resulting in inaccurate imaging isocenter position and geometric distortion. Dependence on gantry position further complicates the correction of distortions. This piece of work presents a new clinical application of a commercial phantom and software arrangement that quantifies isocenter alignment and geometric distortion, as well every bit providing a deformation vector field (DVF). A big distortion phantom and a smaller grid phantom were imaged at multiple gantry angles from 0 to 330° on a 0.35 T integrated MR‐Linac. The software package was used to appraise geometric baloney and generate DVFs to right distortions inside the phantom volume. The DVFs were applied to the filigree phantom with resampling software so evaluated using structural similarity index measure (SSIM). Scans were also performed with a ferromagnetic clip nearly the phantom to investigate the correction of more severe artifacts. The mean magnitude isocenter shift was 0.67 mm, ranging from 0.25 to 1.04 mm across all angles. The DVF had a mean component value of 0.27 ± 0.02, 0.24 ± 0.01, and 0.nineteen ± 0.01 mm in the right‐left (RL), anterior‐posterior (AP), and superior‐junior (SI) directions. The ferromagnetic clip increased isocenter position mistake from 1.98 mm to ii.twenty mm and increased mean DVF component values in the RL and AP directions. The resampled grid phantom had an increased SSIM for all gantry angles compared to original images, increasing from 0.26 ± 0.001 to 0.lxx ± 0.004. Through this clinical assessment, we were able to right geometric distortion and isocenter shift related to gantry position on a 0.35 T MR‐Linac using the distortion phantom and software package. This provides encouragement that information technology could be used for quality assurance and clinically to right systematic baloney caused by imaging at dissimilar gantry angles.
Keywords: distortion correction, MR‐Linac, MRgRT
1. INTRODUCTION
Magnetic resonance guided radiotherapy (MRgRT) has get an important treatment modality for improving treatment outcomes by increasing local command rates and reducing normal tissue toxicity. 1 , 2 , three Systems that combine a magnetic resonance imaging (MRI) scanner and a linear accelerator (Linac) provide axle gating, real‐fourth dimension target tracking, and the ability to perform online adaptive radiotherapy. 3 , four , 5 Additionally, soft tissue dissimilarity and structure identification with on‐board MRI systems are superior to that provided by on‐board X‐ray systems, both for 2D planar images and volumetric images such as cone beam CT (CBCT), without the improver of any excess ionizing radiation. half-dozen , seven
Every advantage provided by MRI guidance is dependent on the spatial and geometric allegiance of the resulting images. MR images take intrinsic distortions which must exist addressed prior to using images for target tracking or adaptive planning to prevent inaccurate dose calculation or improper beam gating. eight , 9 , ten These intrinsic distortions are generated by gradient nonlinearity, main static magnetic field (B0) inhomogeneity, imperfect shimming, and eddy currents, and manifest every bit distorted patient surface, spatial shift of the imaging volume, and distortion of internal structures. viii , 11 , 12 , 13 , fourteen On an MR‐Linac system additional sources of distortion are introduced by the radiotherapy system due to imperfect radiofrequency shielding for Linac components and the presence of ferromagnetic material around the MRI bore. Hu et al. provided a comprehensive QA program for characterizing imaging operation on the Cobalt‐60 based ViewRay MRgRT system. 15 They measured spatial integrity at a single gantry bending and found an average geometric distortion of i.5 and 2.seven mm at 20 and 35 cm diameter spherical volumes (DSV), respectively, when compared to CT marker positions. Additionally, magnetic field homogeneity at five gantry angles was measured, showing that over a 45 cm, DSV the peak‐to‐peak variation was less than 25 ppm at all measured angles. Additional studies at multiple gantry angles accept shown that the position of the Linac gantry can impact the imaging isocenter position by virtually two mm and deform rigid bodies over the unabridged field of view. xvi , 17 , xviii For clinical imaging with the ViewRay MR‐Linac organisation the manufacturer recommends using a unmarried home gantry position for acquiring all setup and planning images, where the imaging organization is tuned at commissioning. Comprehensive MRgRT QA protocols accept been developed on other systems, such equally the i.v T Elekta Unity MR‐Linac system. 19 The work past Tijssen et al. showed a varying level of gantry angle dependency for B0 field homogeneity across four Elekta Unity systems and four gantry angles, with ane system showing no dependency and another displaying increased inhomogeneity with active shimming in place.
Nosotros previously reported the bear upon of gantry position on prototype quality for the ViewRay MRIdian 0.35 T MR‐Linac system (Oakwood Village, OH) using a Fluke 76‐907 uniformity and linearity phantom (HP Manufacturing, Cleveland, OH), a proprietary software provided by ViewRay for spatial integrity analysis, and imaging sequences provided past ViewRay. While this written report produced valuable information, it relied on assessing machine operation with software provided by the same company. Boosted studies take been performed using in‐house volumetric phantoms to assess distortion on low‐field MR images at multiple radiotherapy centers, for a limited number of gantry angles. xx , 21 , 22 However, the in‐house phantoms take been heavy and complicated to properly prepare upwardly. A lightweight spatial integrity phantom and assay software produced past Modus Medical Devices Inc. (Modus QA) (London, Ontario, Canada) offers the possibility to examine the imaging distortion with a tertiary‐political party system. In addition to assessing image distortion, the phantom can be used to generate deformation vector fields (DVFs) to correct systematic distortions over a big field of view.
This written report presents the first clinical feel with the Modus QA QUASAR™ MRID3D geometric distortion phantom and software organisation to mensurate the spatial distortion and imaging isocenter shift related to gantry position on an MR‐Linac arrangement. Using a deformation vector field generated by the software system from the MRID3D geometric distortion phantom, we also corrected the distorted image back to the original geometry, providing the possibility for systematic distortion correction of MR images. DVFs were as well applied to an independent phantom at different gantry angles and to the MRID3D geometric distortion phantom in the presence of metal antiquity distortion to assess systematic corrections.
2. METHODS
2.1. Spatial integrity phantoms
This study used two phantoms that were both imaged on a 0.35 T ViewRay MRIdian MR‐Linac system in MRI QA mode. 3D slicewise images were acquired with the gantry in 12 different positions from 0° to 330° in 30° increments. The first phantom imaged was the QUASAR™ MRID3D cylindrical phantom (Modus QA, London, Ontario, Canada). This phantom has a diameter of 394 mm and a length of 391 mm. The phantom interior is a 25 L air‐filled space, with a closed surface containing 1502 machined fiducials containing paraffinic mineral oil. Fiducials are 6 mm long and have a bore of 3 mm, with the exception of vi positioning fiducials that are 5 mm in diameter. All fiducials are spaced at eighteen mm increments. This conquering was repeated for the MRID3D cylindrical phantom over two separate sessions (S1 and S2). The second phantom was a 3D filigree validation phantom provided by Modus QA. This cylindrical phantom incorporates regularly spaced acrylic grids into the fundamental mineral oil‐filled volume. The grid portion of the phantom is 150 mm across and 150 mm long, with each grid voxel containing a volume of 15 × 15 × 15 mm3. A phantom holder that was built in‐house was used to hold the phantom in place. Both phantoms are shown in Figureane. The same clinical TRUFI sequence was used for each phantom with TR/TE: 3.0/1.0 ms, flip angle: 60°, FoV: 450 × 450 × 360 mm3, acquisition voxel size: ane.five × 1.5 × 1.5 mmthree, readout BW: 570 Hz/Px. Images were also acquired with the on‐board scanner distortion correction role turned on (DstOn) and off (DstOff) and the stage encoding direction applied in both the anterior‐posterior (AP) and posterior‐anterior directions, according to manufacturer instruction. Effigy2 displays centric and coronal views of the QUASAR™ MRID3D cylindrical and 3D grid validation phantoms with DstOn and DstOff.
Images of the (a) QUASARTM MRID3D, (b) Modus QA filigree validation phantom, and (c) the Fluke 76‐907 Uniformity and Linearity phantoms are shown during setup on the 0.35T MR‐Linac
Centric and coronal views of all phantoms referenced in this written report. The QUASARTM MRID3D cylindrical phantom is shown with the on‐board baloney correction turned on (a) and off (b). The Modus QA filigree validation phantom is shown with the on‐board distortion correction turned on (c) and off (d). The Fluke 76‐907 Uniformity and Linearity phantom that was used in previous studies
The geometric baloney and isocenter shift values were compared to previous piece of work by this group in which they were measured for multiple gantry angles using a 2nd spatial integrity phantom, the 2d Fluke 76‐907 Uniformity, and Linearity water phantom. The 2d spatial integrity phantom contains a plane of marker posts and must be setup in at least three orientations to acquire sufficient data. This phantom is used forth with the Spatial Integrity Analysis second software provided by ViewRay for quality balls. The phantom is shown in Figure1, and MR images in the axial and coronal planes are shown in Figure2.
An boosted set of distortion scans was completed at a gantry angle of 300°. For these images, a 33 mm long ferromagnetic clip (i.eastward., paperclip) was placed in a plastic container and bound with paper packing material. The plastic container was so secured to the treatment table in the same centric plane as the imaging isocenter using tape. The container was in the same position for each phantom and the treatment table was returned to the same absolute position readout after switching between phantoms. The aforementioned browse parameters were used for these acquisitions, with the ferromagnetic clip in place and removed from the room, and the baloney correction was too set to on and off.
2.2. Geometric distortion analysis and correction
MRID3D phantom images were uploaded to the QUASAR™ MRID3D geometric distortion analysis software. The software produced principal component fault values based on harmonic analysis of the fiducials as well every bit imaging isocenter alignment. All isocenter shifts are relative to gantry angle of 300°. geometric distortion and isocenter shift values were recorded for the entire phantom. DVFs were then exported for all scans. The exported DVFs were applied to the respective 3D grid validation phantom images using prototype Arrangement distortion resampler software provided past Modus QA. Resampled images were then compared to the original and resampled images acquired at gantry angle of 0° where the system baseline shimming was ready during a system upgrade. Structural similarity index mensurate (SSIM) was calculated in MATLAB 2020b (MathWorks, Natick, MA).
iii. RESULTS
3.i. Gantry dependent geometric distortion
Volumetric 3D images were acquired in ane orientation for the QUASAR™ MRID3D cylindrical phantom due to its big field of view. Figure3 shows the mean (±ane SD) DVF components past gantry angle for images acquired with DstOff (Figure 3a) and DstOn (Figure 3b). Boilerplate DVF values at gantry angle of 300° from the two imaging sessions for DstOff scans were ii.12 ± 0.01, 0.99 ± 0.02, and 2.29 ± 0.01 mm in the anterior‐posterior (AP), superior‐inferior (SI), and correct‐left (RL) directions, respectively. Average DVF values at gantry bending of 300° from the ii imaging sessions for DstOn scans were 0.24 ± 0.01, 0.nineteen ± 0.01, and 0.27 ± 0.02 mm in the AP, SI, and RL directions, respectively. The maximum DVF values at gantry angle of 300° for the DstOff scans were 15.26, five.71, and 15.32 mm, in the AP, SI, and RL directions, respectively, and were 1.61, one.46, and i.xl mm in the AP, SI, and RL directions for the DstOn scans.
Mean distortion correction values for the right‐left (RL), inductive‐posterior (AP), and superior‐inferior (SI) directions beyond all gantry angles for imaging sessions 1 and two, with the stage encoding management in the anterior‐posterior (AP) and posterior‐anterior (PA) directions. Plots show the values with (a) distortion correction off (DstOff) and with (b) distortion correction on (DstOn)
Isocenter shift values for each gantry angle are shown in Figure4, in the AP, SI, and RL directions, equally well every bit the magnitude shift value. The maximum magnitude isocenter shift of i.04 and 1.02 mm for DstOn and DstOff, respectively, occurred at a gantry angle of 120° for the AP phase encoding direction. The PA phase encoding direction had a maximum magnitude isocenter shift of i.01 and one.02 mm for DstOn and DstOff, respectively, at a gantry angle of 90°.
Isocenter distance to agreement (DTA) for the (a) magnitude, (b) correct‐left (RL), (c) anterior‐posterior (AP), and (d) superior‐inferior (SI) directions. Each plot includes the values from both imaging sessions with the phase encoding management in the inductive‐posterior (AP) and posterior‐anterior (PA) directions, and with the distortion correction turned on (DstOn) and off (DstOff), likewise as the previous data from Lewis et al18
In guild to assess the precision of the isocenter shift measurements, the arrangement baloney resampler software was used to generate copies of an AP phase encoding direction DstOn scan, shifted in the y‐management past amounts ranging from 0.05 to 1.5 mm. The detected centroid values from these images were compared to the expected position, which is the detected position in the original image plus the applied shift. The resulting measurement error ranged from −0.05 to 0.09 mm. A comparison of sets of four series caused on the same twenty-four hours and at the same gantry angle showed that the maximum variation in the in‐plane phase encoding component of the centroid within each group was 0.06 mm.
iii.2. Distortion correction
3.ii.1. Gantry position
Application of the DVFs from the MRID3D cylindrical phantom distortion measurement was successfully applied to the grid phantom for resampling. Effigyv shows the filigree phantom pre‐ and post‐correction for ii gantry angles (120° and 300°). The field of view is cut off due to the resampler just applying the DVF to the volume encompassed by the MRID3D cylindrical phantom.
The grid phantom acquired at gantry angle 120° (a) and 300° (b), showing the original images and the resampled images using the appropriate deformation vector field generated using the QUASARTM MRID3D geometric distortion analysis software, and an overlayed paradigm showing matching pixels in grayscale. The scale bar indicates the 150 mm length
Resampling increased the average SSIM value for all gantry angles relative to resampled gantry 0° images from 0.26 ± 0.001 to 0.06 ± 0.00 prior to resampling, from 0.lxx ± 0.004 to 0.69 ± 0.005 post resampling for DstOn and DstOff, respectively. SSIM was evaluated against the resampled gantry 0° epitome due to slight blurring of structures after resampling. Figurevi displays SSIM values for DstOn and DstOff images at all gantry angles, pre‐ and post‐resampling.
SSIM values for all measured gantry angles pre‐ and post‐resampling. Shown with the onboard distortion correction role turned on (a) and turned off (b)
3.2.2. Metal object in‐field
With the ferromagnetic clip in place, the geometric baloney of the MRID3D cylindrical phantom was increased from 0.27 ± 0.31 to 0.36 ± 0.44 mm in the RL direction and from 0.24 ± 0.34 to 0.27 ± 0.36 mm in AP management. The correction remained abiding in the SI direction at 0.29 ± 0.33 mm. The imaging isocenter distance to agreement increased from a magnitude of 1.98 to 2.20 mm with the paperclip in place. Slight distortions were visible near the border of the field. The DVF was then applied to the grid phantom using the system distortion resampler software. The SSIM value between the resampled grid phantom without the ferromagnetic clip and the grid phantom with the ferromagnetic clip was 0.07 pre‐resampling and was 0.65 post‐resampling. The QUASAR™ MRID3D cylindrical phantom, and pre‐ and postal service‐correction filigree phantom images with the paperclip in identify are displayed in Figure7.
The QUASARTM MRID3D phantom (a), Modus QA grid validation phantom pre‐resampling (b), and the Modus QA grid phantom post‐resampling (c) with a ferromagnetic clip in place on the treatment table in the same centric aeroplane as the image. The blueish arrows betoken regions of baloney acquired past the ferromagnetic clip in all 3 images. The scale bar indicates a 150 mm length for each image
4. Give-and-take
This study presents the starting time clinical experience with the gantry‐related prototype distortion correction using the QUASAR™ MRID3D cylindrical phantom and the associated geometric baloney analysis software on an institutional 0.35 T MR‐Linac system. This phantom‐based approach to measure out geometric distortion and isocenter shift allowed for a vendor independent alternative to the current clinical practice that utilizes software provided by ViewRay to perform quality balls on a ViewRay machine. This organisation also uses a large 3D phantom which covers a big imaging field of view. Measuring distortion over a big field of view is valuable for assessing distortion near the periphery of the imaging region without the need for repositioning the phantom and acquiring additional image sets. The geometric distortion and isocenter shift were quantified at multiple gantry angles, additionally, big field of view DVFs were produced to correct the systematic distortions nowadays in caused MR images. Comparison of isocenter shifts to previous data published by Lewis et al., xviii which used the Fluke spatial integrity phantom, showed practiced agreement with the maximum isocenter shift of 0.89 mm occurring at gantry angle of 120°, equally it did in this work. The calculated DVF allowed for an independent smaller filigree phantom to be corrected back to its original geometry over multiple gantry angles. This gives the possibility for gantry bending‐specific systematic distortion corrections to be applied to clinical images. Geometric correction of clinical images at multiple gantry angles could reduce imaging time and the demand for a set gantry imaging position when re‐acquisition of 3D anatomical images is required during treatment. Additionally, the software was able to correct for a minor metallic artifact created past a paperclip at the edge of the field. This indicates that if unlike systematic distortions are nowadays on different imaging systems, that the software could correct them. However, information technology cannot be used to right for a patient‐specific metallic artifact because it requires a reference template.
MRI isocenter shift results showed that there is a systematic change in isocenter position with gantry angle. The results in this written report closely agree with previous works which used unlike phantoms and analysis techniques. 16 , xviii This piece of work as well plant that the on‐board distortion correction was able to right the geometric baloney to a similar value at all gantry angles, and was a significant comeback over images with the distortion correction turned off. Gantry angle of 0° showed a college distortion correction value in the AP direction for DstOn scans than other gantry angles, simply was the distortion was however corrected past the software.
Hereafter work with this system will be directed at systematic distortion correction of homo images acquired at multiple gantry angles. However, this extension presents many challenges, including assessing correction accurateness and the long acquisition time for imaging at multiple gantry angles. The phantom provides a large field of view, but it may not be sufficient to capture a human discipline which could result in an incomplete distortion correction.
This work demonstrated the ability to perform phantom‐based paradigm baloney correction via quantifying and applying deformation vectors to a large cylindrical phantom in a clinical setting and provided a reliable alternative to current imaging quality balls procedures on integrated MR‐Linac systems.
five. Conclusion
The first clinical experience with a new geometric distortion analysis and correction organization was presented in this work. This organization showed proficient agreement in imaging isocenter and image distortion values every bit seen in previous piece of work, with imaging isocenter position varying depending on gantry bending and having a maximum departure of 1.04 mm, with the distortion correction on. Large volume DVFs could as well be applied to correct systematic geometric baloney in a secondary phantom, even in the presence of metallic‐induced artifacts.
CONFLICT OF INTEREST
The authors declare no conflicts of interest in this work.
Author CONTRIBUTIONS
Benjamin Lewis contributed to information drove, analysis, and wrote the manuscript. Jaeik Shin contributed to data collection. Benjamin Quinn and Enzo Barberi contributed to pattern of information analysis, analysis tools, and provided technical assist. Domenic Sievert contributed to data collection and analysis. Jin Sung Kim contributed to study design and supervision. Taeho Kim contributed to data collection, analysis, and supervision of the project. All authors discussed the results and contributed to the final manuscript.
ACKNOWLEDGEMENTS
The Department of Radiation Oncology provided the inquiry funding for this piece of work to Dr. Taeho Kim. Washington University in St. Louis has a chief research agreement and receives research funding from ViewRay unrelated to this study. Dr. Jin Sung Kim was supported by the National Inquiry Foundation of Korea (NRF) grant funded past the Korean Government (MSIT) (No. 2020R1A4A101661911).
Notes
Lewis BC, Shin J, Quinn B, et al. First clinical experience of correcting phantom‐based image baloney related to gantry position on a 0.35 T MR‐Linac. J Appl Clin Med Phys. 2021;22(xi):21–28. 10.1002/acm2.13404 [PMC free commodity] [PubMed] [CrossRef] [Google Scholar]
DATA AVAILABILITY STATEMENT
The data that support the findings of this report are available from the respective author upon reasonable asking.
REFERENCES
1. Henke L, Kashani R, Yang D, et al. Adaptive MR‐guided stereotactic torso radiations therapy (AMR‐SBRT) for oligometastatic or unresectable primary abdominal malignancies: results of a prospective stage I trial. Int J Radiat Oncol Biol Phys. 2016;96(2):E205‐E206. 10.1016/j.ijrobp.2016.06.1108 [CrossRef] [Google Scholar]
2. Henke Fifty, Kashani R, Robinson C, et al. Phase I trial of stereotactic MR‐guided online adaptive radiations therapy (SMART) for the treatment of oligometastatic or unresectable primary malignancies of the abdomen. Radiother Oncol. 2018;126(3):519‐526. ten.1016/j.radonc.2017.eleven.032 [PubMed] [CrossRef] [Google Scholar]
3. Rudra S, Jiang N, Rosenberg SA, et al. Using adaptive magnetic resonance image‐guided radiations therapy for treatment of inoperable pancreatic cancer. Cancer Med. 2019;8(5):2123‐2132. 10.1002/cam4.2100 [PMC costless article] [PubMed] [CrossRef] [Google Scholar]
4. Green OL, Rankine LJ, Cai B, et al. Offset clinical implementation of real‐time, real anatomy tracking and radiation axle control. Med Phys. 2018;45(viii):3728‐3740. ten.1002/mp.13002 [PubMed] [CrossRef] [Google Scholar]
5. van Sörnsen de Koste JR, Palacios MA, Bruynzeel AME, Slotman BJ, Senan S, Lagerwaard FJ. MR‐guided gated stereotactic radiation therapy commitment for lung, adrenal, and pancreatic tumors: a geometric analysis. Int J Radiat Oncol Biol Phys. 2018;102(4):858‐866. 10.1016/j.ijrobp.2018.05.048 [PubMed] [CrossRef] [Google Scholar]
6. Fischer‐Valuck BW, Henke 50, Light-green O, et al. Two‐and‐a‐half‐year clinical experience with the world's first magnetic resonance prototype guided radiation therapy arrangement. Adv Radiat Oncol. 2017;two(3):485‐493. x.1016/j.adro.2017.05.006 [PMC free commodity] [PubMed] [CrossRef] [Google Scholar]
7. Cerviño LI, Du J, Jiang SB. MRI‐guided tumor tracking in lung cancer radiotherapy. Phys Med Biol. 2011;56(13):3773‐3785. 10.1088/0031-9155/56/13/003 [PubMed] [CrossRef] [Google Scholar]
viii. Spees WM, Buhl N, Dominicus P, Ackerman JJ, Neil JJ, Garbow JR. Quantification and compensation of eddy‐current‐induced magnetic‐field gradients. J Magn Reson. 2011;212(1):116‐123. 10.1016/j.jmr.2011.06.016 [PMC gratuitous article] [PubMed] [CrossRef] [Google Scholar]
nine. Yan Y, Yang J, Beddar Due south, et al. A methodology to investigate the affect of image distortions on the radiation dose when using magnetic resonance images for planning. Phys Med Biol. 2018;63(8):085005. ten.1088/1361-6560/aab5c3 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
x. Seibert TM, White NS, Kim GY, et al. Distortion inherent to magnetic resonance imaging can lead to geometric miss in radiosurgery planning. Pract Radiat Oncol. 2016;6(half-dozen):e319‐e328. ten.1016/j.prro.2016.05.008 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
11. Huang SY, Seethamraju RT, Patel P, Hahn PF, Kirsch JE, Guimaraes AR. Body MR imaging: artifacts, one thousand‐space, and solutions‐erratum. Radiographics. 2015;35(5):1624. 10.1148/rg.2015154016 [PMC gratuitous article] [PubMed] [CrossRef] [Google Scholar]
12. Walker A, Liney Yard, Holloway L, Dowling J, Rivest‐Henault D, Metcalfe P. Continuous tabular array conquering MRI for radiotherapy treatment planning: distortion assessment with a new extended 3D volumetric phantom. Med Phys. 2015;42(four):1982‐1991. 10.1118/1.4915920 [PubMed] [CrossRef] [Google Scholar]
13. Baldwin LN, Wachowicz K, Thomas SD, Rivest R, Fallone BG. Characterization, prediction, and correction of geometric distortion in 3 T MR images. Med Phys. 2007;34(two):388‐399. 10.1118/1.2402331 [PubMed] [CrossRef] [Google Scholar]
14. Weavers PT, Tao Southward, Trzasko JD, et al. Epitome‐based gradient not‐linearity characterization to determine higher‐social club spherical harmonic coefficients for improved spatial position accuracy in magnetic resonance imaging. Magn Reson Imaging. 2017;38:54‐62. 10.1016/j.mri.2016.12.020 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
15. Hu Y, Rankine 50, Green OL, et al. Characterization of the onboard imaging unit for the outset clinical magnetic resonance image guided radiation therapy system. Med Phys. 2015;42(10):5828‐5837. x.1118/1.4930249 [PubMed] [CrossRef] [Google Scholar]
16. Kim T, Gu B, Maraghechi B, et al. Characterizing MR imaging isocenter variation in MRgRT. Biomed Phys Eng Limited. 2020;6(3):035009. 10.1088/2057-1976/ab7bc6 [PubMed] [CrossRef] [Google Scholar]
17. Mittauer KE, Dunkerley DAP, Yadav P, Bayouth JE. Label and longitudinal assessment of daily quality assurance for an MR‐guided radiotherapy (MRgRT) Linac. J Appl Clin Med Phys. 2019;20(11):27‐36. 10.1002/acm2.12735 [PMC gratuitous article] [PubMed] [CrossRef] [Google Scholar]
18. Lewis BC, Gu B, Klett R, Lotey R, Light-green OL, Kim T. Label of radiotherapy component impact on MR imaging quality for an MRgRT system. J Appl Clin Med Phys. 2020;21(12):20‐26. ten.1002/acm2.13054 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
nineteen. Tijssen RHN, Philippens MEP, Paulson ES, et al. MRI commissioning of one.5T MR‐Linac systems – a multi‐institutional study. Radiother Oncol. 2019;132:114‐120. 10.1016/j.radonc.2018.12.011 [PubMed] [CrossRef] [Google Scholar]
20. Ginn JS, Agazaryan N, Cao M, et al. Characterization of spatial distortion in a 0.35 T MRI‐guided radiotherapy arrangement. Phys Med Biol. 2017;62(11):4525‐4540. [PMC gratis article] [PubMed] [Google Scholar]
21. Huang KC, Cao Y, Baharom U, Balter JM. Phantom‐based characterization of distortion on a magnetic resonance imaging simulator for radiation oncology. Phys Med Biol. 2016;61(2):774‐790. [PMC free article] [PubMed] [Google Scholar]
22. Nejad‐Davarani SP, Kim JP, Du D, Glide‐Hurst C. Large field of view distortion cess in a depression‐field MR‐Linac. Med Phys. 2019;46(5):2347‐2355. ten.1002/mp.13467 [PMC gratuitous article] [PubMed] [CrossRef] [Google Scholar]
Articles from Journal of Practical Clinical Medical Physics are provided here courtesy of Wiley-Blackwell
Source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8598146/
Post a Comment for "First Take a Picture First Before Uploading Phantom"