Value of Automated Coronal Reformations From 64-Section Multidetector Row Computerized Tomography in the Diagnosis of Urinary Stone Disease

Autor: Raul N. Uppot, Wen-Chiung Lin, Chao-Shiang Li, Peter F. Hahn, Dushyant V. Sahani
Rok vydání: 2007
Předmět:
Zdroj: Journal of Urology. 178:907-911
ISSN: 1527-3792
0022-5347
DOI: 10.1016/j.juro.2007.05.042
Popis: We determined the value of automated coronal reformation using 64-detector computerized tomography for the detection of urinary stones.A total of 72 patients underwent unenhanced 64-detector computerized tomography to diagnose urinary stones. Two radiologists independently reviewed coronal reformations and axial images at separate reading sessions. The stone detection rate, reader confidence and interpretation time per radiologist were recorded. Two radiologists reviewed coronal and axial images in consensus and served as the reference standard.A total of 175 stones were diagnosed by consensus. Using coronal reformations 162 stones (92.6%) were detected by reader 1 and 157 (89.7%) were detected by reader 2. Using axial images 157 stones (90.3%) were detected by reader 1 and 155 (88.6%) were detected by reader 2. The reading time of coronal reformations was significantly shorter than that of axial images for each reader (p0.01). Using coronal imaging to complement axial imaging 12 additional stones were detected and 23 were diagnosed with increased confidence by reader 1, while an additional 15 were detected and 8 were diagnosed with increased confidence by reader 2. The mean size of stones detected with coronal reformations alone was significantly smaller than that of the total stones. Excellent interobserver agreement was noted for coronal reformations and axial images (kappa coefficient: 0.91 and 0.904, respectively).Review of automated coronal reformations allows equally accurate and more rapid detection of urinary stones compared with axial images alone. In addition, coronal reformation of 64-detector computerized tomography adds value when used in conjunction with axial data sets.
Databáze: OpenAIRE