Premium
DWI of the prostate: Comparison of a faster diagonal acquisition to standard three‐scan trace acquisition
Author(s) -
CorcueraSolano Idoia,
Wagner Mathilde,
Hectors Stefanie,
Lewis Sara,
Titelbaum Nicholas,
Stemmer Alto,
Rastinehad Ardeshir,
Tewari Ashutosh,
Taouli Bachir
Publication year - 2017
Publication title -
journal of magnetic resonance imaging
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.563
H-Index - 160
eISSN - 1522-2586
pISSN - 1053-1807
DOI - 10.1002/jmri.25705
Subject(s) - effective diffusion coefficient , mcnemar's test , medicine , nuclear medicine , reproducibility , magnetic resonance imaging , prostate cancer , diffusion mri , image quality , prostate , wilcoxon signed rank test , radiology , mann–whitney u test , cancer , mathematics , artificial intelligence , image (mathematics) , statistics , computer science
Purpose To compare a faster diagonal diffusion‐weighted imaging (d‐DWI) to conventional three‐scan trace DWI (t‐DWI) acquisition for prostate magnetic resonance imaging (MRI) in terms of image quality, tumor detection/conspicuity, Prostate Imaging Reporting and Data System (PI‐RADS) characterization, quantitative estimated signal‐to‐noise ratio (eSNR), and apparent diffusion coefficient (ADC) measurement. Patients and Methods A total of 34 consecutive men with suspected prostate cancer (PCa) who underwent 3T MRI of the prostate were assessed. MRI included t‐DWI and d‐DWI (using b‐values of 50, 1000, and 1600 s/mm 2 , number of averages 1/5/10 for t‐DWI vs. 2/8/14 for d‐DWI, acquisition time 6:21 min vs. 4:17 min, respectively). Two independent observers evaluated image quality, including image sharpness, anatomic distortion, and artifacts on a 5‐point scale (1–5) and assessed tumor detection, conspicuity, and PI‐RADS classification with both DWI sequences. eSNR and ADC were measured in the peripheral zone (PZ), transitional zone (TZ), and detected tumors. Data was compared using paired Wilcoxon signed rank tests and McNemar test. Coefficients of variations (CV) between ADC obtained with both sequences were calculated. Results Significantly fewer artifacts were observed on d‐DWI at b 1600 for observer 2 ( P < 0.01), while the other image quality scores were equivalent for both sequences. eSNR was lower with d‐DWI vs. t‐DWI in PZ and TZ for b 1000 ( P < 0.01 and P = 0.03, respectively) and b 1600 ( P < 0.01 for both). ADC reproducibility between sequences was excellent (CV <10%). No significant differences were found for tumor detection ( P ≥ 0.25), conspicuity ( P ≥ 0.12), or PI‐RADS classification ( P ≥ 0.10). Conclusion d‐DWI can provide a substantial reduction in acquisition time (∼30%) while maintaining equivalent tumor detection, PI‐RADS score, image quality, and ADC values. Level of Evidence: 3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2017;46:1767–1775.