(Sunday, 7/29/2018) 3:00 PM - 6:00 PM
Room: Exhibit Hall
Purpose: Metal artifacts in computed tomography CT images are one of the main problems in radiation oncology as they introduce uncertainties to target and organ at risk delineation as well as dose calculation. This study evaluated the performance of the smart metal artifact reduction function for radiation therapy from General Electric and investigated its clinical impact in treatment planning.
Methods: In a phantom study the CT artifacts caused by metals with different densities: aluminum (Ï? Al=2.7 g/cm(3)), titanium (Ï? Ti=4.5 g/cm(3)) and steel (Ï? steel=7.9 g/cm(3)) have been investigated. Data were collected using a clinical scanner (General Electric 580 RT). A data set without metal substitute served as the reference. The CT Hounsfield numbers of density plugs on both uncorrected and SMAR corrected images were compared. Treatment planning accuracy was evaluated by comparing simulated dose distributions computed using the true density images, uncorrected images, and SMAR corrected images.
Results: Results of the phantom study indicated that CT Hounsfield number accuracy and noise were improved on the SMAR corrected images, especially for images with bilateral metal implants. The average deviation of HU decreased from 1035 HU to 496 HU in areas including metal and from 352 HU to 23 HU in tissue areas excluding metal. Dose calculation accuracy could be significantly improved for all phantoms and plans. The mean passing rate for gamma evaluation with 3% dose tolerance and 3 mm distance to agreement increased from 91.3% to 98.4% if artifacts were corrected by SMAR.
Conclusion: The application of SMAR allows metal artifacts to be removed to a great extent which leads to a significant increase in dose calculation accuracy. However, MAR may also introduce new image artifacts especially when used on titanium implants. Therefore, it is important to reconstruct CT images containing metal with and without MAR.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
