PURPOSE: To improve spatial resolution and image quality of diffusion-weighted (DW) MRI in detecting low-risk prostate cancer (lrPC) in patients undergoing active surveillance protocol (AS-PC), we propose the application of a diffusion-prepared balanced steady-state free precession (bSSFP) technique capable of multishot acquisition.
METHODS: Diffusion-prepared bSSFP was compared with single-shot DW echo planar imaging (SS-DW-EPI) at two prescribed resolutions (2.1 × 2.1 × 3.5mm3 , 0.9 × 0.9 × 3.5 mm3 ) in nine healthy subjects and nine AS-PC patients. Geometric distortion and susceptibility artifacts were quantitatively assessed in all subjects. In AS-PC patients, lesion detection via blinded multiparametric MRI including T1-weighted, T2-weighted, dynamic contrast-enhanced imaging, and along with either of two DW methods were evaluated against 12-point biopsy.
RESULTS: Geometric distortion and susceptibility artifacts were significantly less for diffusion-prepared bSSFP at both prescribed spatial resolutions than SS-DW-EPI. Apparent diffusion coefficients of healthy prostate tissue were concordant between the two DW methods at both spatial resolutions. In AS-PC patients, multiparametric MRI with diffusion-prepared bSSFP had greater sensitivity (94%, 63%), accuracy (76%, 67%), positive-predictive value (54%, 48%), negative-predictive value (97%, 82%), and area under the curve (0.80, 0.67) than with SS-DW-EPI.
CONCLUSIONS: The proposed diffusion-prepared technique with higher spatial resolution and improved image quality over SS-DW-EPI resulted in better multiparametric MRI detection of lrPC in AS-PC patients.
Written by:
Nguyen C, Sharif-Afshar AR, Fan Z, Xie Y, Wilson S, Bi X, Payor L, Saouaf R, Kim H, Li D. Are you the author?
Biomedical Imaging Research Institute, Cedars-Sinai Medical Center, Los Angeles, California, USA; Department of Bioengineering, University of California Los Angeles, Los Angeles, California, USA.
Reference: Magn Reson Med. 2015 Mar 11. Epub ahead of print.
doi: 10.1002/mrm.25609
PubMed Abstract
PMID: 25761871