Meeting Banner
Abstract #1336

Deep learning for synthesizing high-b-value DWI of the prostate: A tentative study based on generative adversarial networks

lei hu1, jungong Zhao1,2, Caixia fu3, and Thomas Benkert4
1Department of Diagnostic and Interventional Radiology, Shanghai Jiao Tong University Affiliated Sixt, 上海, China, 2Department of Diagnostic and Interventional Radiology, Shanghai Jiao Tong University Affiliated Sixt, shanghai, China, 3MR Application Development, Siemens Shenzhen magnetic Resonance Ltd, shanghai, China, 4MR Application Predevelopment, Siemens Healthcare, Erlangen, Gernmany, Erlangen, Germany

A deep learning framework based on a generative adversarial network (GAN) to synthesize high-b-value DWI (syn-DWIb1500) with high quality using the acquired standard b-value DWI (a-DWIb800-1000) was developed. Reader ratings for image quality and PCa detection were performed on the a-DWI b1500, syn-DWIb1500, and optimized syn-DWIb1500 sets. Wilcoxon signed-rank tests and MRMC-ROC were used to compare the readers’ scores and diagnostic capabilities of each DWI set, respectively. Optimized syn-DWIb1500 resulted in significantly better image quality (all P≤0.001) and a higher mean AUC than a-DWIb1500 and cal-DWIb1500 (all P≤0.042).

This abstract and the presentation materials are available to members only; a login is required.

Join Here

Abstract