| Characterization of urinary stones with dual-energy CT: improved differentiation using a tin filter. | |
| | |
MedLine Citation:
|
PMID: 19996763 Owner: NLM Status: MEDLINE |
Abstract/OtherAbstract:
|
PURPOSE: To assess image quality and capability of stone differentiation between UA-containing and non-UA-containing uroliths with the latest dual-energy (DE) computed tomography (CT) system equipped with a tin filter (TF) using various data acquisition parameters in a work bench model. METHODS AND MATERIALS: One hundred ten urinary stones (4.2 +/- 3.0 mm, 0.4-12 mm) of 15 compositions were examined in an ex vivo phantom, using DE dual-source CT (Definition Flash, Siemens Healthcare) equipped with a TF. Phantom was scanned in a water tank and contained stones in acrylic elliptic spheres filled with a parenchyma substitute. Scans were performed at 3 different settings: at 80 and 140 kVp without TF, at 80 and 140 kVp with TF, and at 100 and 140 kVp with TF. Tube current time products were adapted to yield constancy in CT dose indices (CTDIvol = 18.84 mGy, 18.95 mGy, and 18.90 mGy, respectively). CT numbers of urinary stones and image noise were electronically measured by placing regions of interest. DE indices (DEI) were calculated and compared using analysis of variances for repeated measures and paired t tests; image noise (IN) using the Friedman test. The stones were classified as UA-containing or non-UA-containing on color-coded images based on the DEI. Diagnostic accuracy was calculated using crystallographic analysis as standard of reference. RESULTS: Of the 110 stones (60%), 65 contained UA; 45 stones (40%) contained no UA. DEI was greatest at 80 and 140 kVp when using the TF (DEI80 kVp/TF140 kVp = 0.038 vs. DEI80 kVp/140 kVp = 0.028, DEI100 kVp/TF140 kVp = 0.025; P < 0.01). IN of high kVp acquisitions were similar (P = 0.15), whereas IN of low kVp acquisitions were significantly (P < 0.001) different being lowest at 100 kVp. The semiautomated DE software correctly classified all stones at all settings with a diagnostic accuracy of 100% (95% confidence interval: 97%-100%). CONCLUSION: DECT with TF and 80-140 kVp tube voltage settings significantly improves the discrimination between UA-containing and non-UA containing urinary stones as compared with DECT without using the TF on the basis of DEI. The 100/140 kVp setting with TF is associated with lower IN but demonstrates similar discrimination abilities as compared with 80/140 kVp setting without the use of the TF. |
| | |
Authors:
|
Paul Stolzmann; Sebastian Leschka; Hans Scheffel; Katharina Rentsch; Stephan Baum?ller; Lotus Desbiolles; Bernhard Schmidt; Borut Marincek; Hatem Alkadhi |
Related Documents
:
|
9022773 - Efficacy of helical ct in t-staging of gastric cancer. 18197063 - Dual energy ct characterization of urinary calculi: initial in vitro and clinical exper... 10897403 - Inner ear anomaly of three-dimensional computed tomography: computed tomographic attenu... |
Publication Detail:
|
Type: Journal Article |
Journal Detail:
|
Title: Investigative radiology Volume: 45 ISSN: 1536-0210 ISO Abbreviation: Invest Radiol Publication Date: 2010 Jan |
Date Detail:
|
Created Date: 2009-12-22 Completed Date: 2010-03-26 Revised Date: - |
Medline Journal Info:
|
Nlm Unique ID: 0045377 Medline TA: Invest Radiol Country: United States |
Other Details:
|
Languages: eng Pagination: 1-6 Citation Subset: IM |
Affiliation:
|
Institute of Diagnostic Radiology, University Hospital Zurich, Zurich, Switzerland. |
Export Citation:
|
APA/MLA Format Download EndNote Download BibTex |
| MeSH Terms | |
Descriptor/Qualifier:
|
Filtration
/
instrumentation,
methods Humans Kidney Calculi / radiography* Particle Size Phantoms, Imaging Tin* Tomography, X-Ray Computed* Urinary Calculi / radiography* |
| Chemical | |
Reg. No./Substance:
|
7440-31-5/Tin |
From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine
Previous Document: Preliminary Results for Characterization of Pelvic Lymph Nodes in Patients With Prostate Cancer by D...
Next Document: Prediction of HIV type 1 subtype C tropism by genotypic algorithms built from subtype B viruses.