Metallic artifact mitigation and organ-constrained tissue assignment for Monte Carlo calculations of permanent implant lung brachytherapy

Abstract

Despite producing differences in CT images, dose metrics calculated using the STR, fan beam + STR, and 3D median filter techniques produce similar dose metrics. Results suggest that the accuracy of dose distributions for permanent implant lung brachytherapy is improved by applying lung-constrained tissue assignment schemes to metallic artifact corrected images.

Improved tissue assignment using dual-energy computed tomography in low-dose rate prostate brachytherapy for Monte Carlo dose calculation

“…Several researchers have developed different algorithms to remove these artifacts. [20][21][22][23] These approaches use model-based iterative reconstruction (MBIR), with modern CT imaging modalities to correct artifacts within the sinogram. The model corrects for metallic artifacts by subtracting reprojected sinograms of segmented brachytherapy seed images from the raw sinogram.…”

Advancements in Monte Carlo Dose Calculations for Prostate and Breast Permanent Implant Brachytherapy

“…28 Earlier work has shown that the accuracy of dose calculations can be significantly diminished without first implementing a metallic artifact reduction algorithm to post-implant CT images. [29][30][31] Artifact mitigation in CT images has been studied in the context of dental fillings 28,32,33 and prosthesis, [34][35][36][37] and several general approaches including thresholding, 29,30 spatial filters 30,38 and sinogram space filters [29][30][31]39 have been proposed. No published arifact mitigation method has been shown to work perfectly with the artifacts characteristic of brachytherapy images, and it remains unclear how differences in artifact mitigation approaches affect dose calculations.…”

“…This smoothing effect could challenge preservation of soft tissue heterogeneities important for other treatment sites such as breast 47,48 or lung. 29,30 The 3D median filter is one of several spatial-domain filters which could be applied to mitigate artifacts from CT images. We have explored other similar methods, such as Gaussian filters, 2D and 3D filters, varied kernel and clip sizes, and have consistently observed comparative blurring effects.…”

“…MAR methods demonstrated in the literature are mainly for prostheses [34][35][36][37] and dental 28,32,33 applications; their extension to the brachytherapy context may be challenged by the large number of small and tightly spaced seeds, and the different anatomical sites. A few recent studies have begun to explore MAR methods for brachytherapy, demonstrating these techniques in idealized phantoms and a few select clinical cases: approaches include thresholding, 29,30 image space filters, 30,38 and projection space raw sinogram 31,39 and virtual sinogram methods. 29,30 In the prostate, uniformly-distributed calcifications have been shown to significantly decrease the target D 90 , by up to 37% in one study 54 and by up to 10 Gy in another.…”

“…A few recent studies have begun to explore MAR methods for brachytherapy, demonstrating these techniques in idealized phantoms and a few select clinical cases: approaches include thresholding, 29,30 image space filters, 30,38 and projection space raw sinogram 31,39 and virtual sinogram methods. 29,30 In the prostate, uniformly-distributed calcifications have been shown to significantly decrease the target D 90 , by up to 37% in one study 54 and by up to 10 Gy in another. 78 Thus, accurate prostate brachytherapy dose calculations require the modelling of calcifications, however, both calcifications and metallic artifacts may correspond to very high CT numbers, thus challenging the mapping of CT number to tissue type and affecting computed doses.…”