Spectral CT with monochromatic imaging and metal artifacts reduction software for artifacts reduction of 125I radioactive seeds in liver brachytherapy

Purpose To investigate the optimal monochromatic energy for artifacts reduction from 125 I seeds as well as image improvement in the vicinity of seeds on monochromatic images with and without metal artifacts reduction software (MARS) and to compare this with traditional 120-kVp images, so as to eval...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Japanese journal of radiology 2015-11, Vol.33 (11), p.694-705
Hauptverfasser: Yang, Qiuxia, Peng, Sheng, Wu, Jing, Ban, Xiaohua, He, Mingyan, Xie, Chuanmiao, Zhang, Rong
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Purpose To investigate the optimal monochromatic energy for artifacts reduction from 125 I seeds as well as image improvement in the vicinity of seeds on monochromatic images with and without metal artifacts reduction software (MARS) and to compare this with traditional 120-kVp images, so as to evaluate the application value of gemstone spectral imaging for reducing artifacts from 125 I seeds in liver brachytherapy. Materials and methods A total of 45 tumors from 25 patients treated with 125 I seed brachytherapy in the liver were enrolled in this study. Multiphasic spectral computed tomography (CT) scanning was performed for each patient. After a delay time of 15 s of portal vein phase, a traditional 120-kVp scan was performed, focusing on several planes of 125 I seeds only. The artifact index (AI) in the vicinity of seeds and the standard deviation (SD) of the CT density of region of interest in the outside liver parenchyma were calculated. Artifact appearance was evaluated and classified on reconstructed monochromatic S and 120-kVp images. Image quality in the vicinity of seeds of three data sets were evaluated using a 1–5 scale scoring method. The Friedman rank-sum test was used to estimate the scoring results of image quality. Results The greatest noise in monochromatic images was found at 40 keV (SD = 27.38, AI = 206.40). The optimal monochromatic energy was found at 75 keV, which provided almost the least image noise (SD = 10.01) and good performance in artifact reduction (AI = 102.73). Image noise and AI reduction at 75 keV was decreased by 63.44 and 50.23 %, compared with at 40 keV. Near-field thick artifacts were obvious in all 45 lesions, in 120-kVp images, and 75-keV images, but basically reduced in 75 keV MARS images and artifacts completely invisible in 7 lesions. The number of diagnosable images (score ≥3) was significantly more in the 75-keV MARS group (28/45), and the 75-keV group (22/45) than in the 120-kVp group (11/45) ( p  
ISSN:1867-1071
1867-108X
DOI:10.1007/s11604-015-0482-6