Abstract #2389

Learning how to adapt T2 PROPELLER MR prostate imaging: going beyond PIRADS requirements with MR Deep Learning reconstruction

Julie Poujol¹, Charline Henry¹, Vincent Barrau², François Legou², Eric Pessis², Xinzeng Wang³, and Daniel Litwiller⁴

¹Clinical Research & Development, GE Healthcare, Buc, France, ²Centre Cardiologique du Nord, Saint-Denis, France, ³Global MR Applications & Workflow, GE Healthcare, Houston, TX, United States, ⁴Global MR Applications & Workflow, GE Healthcare, New York, NY, United States

To give a confident image-based prostate cancer diagnosis, PIRADS recommends using multiparametric MR (mp-MRI) exam composed of DWI and T2w sequences. By using a new deep learning-based image reconstruction algorithm, we aim to improve the utility of T2w PROPELLER images by reducing acquisition time and/or increasing spatial resolution beyond PIRADS requirements. We present quantitative analysis based on signal-to-noise ratio estimates and qualitative analysis based on delineation of anatomical structures, overall image quality, vision of thin structures.

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

Join Here