Meeting Banner
Abstract #2603

Deep learning based image reconstruction for T2-weighted rectal cancer imaging

Ken-Pin Hwang1, Xinzeng Wang2, Randy D Ernst3, Sarah M Palmquist3, Marc Lebel2, Gaiane M Rauch3, George J Chang4, Craig Messick4, Melissa W Taggart5, Ersin Bayram2, Jingfei Ma1, and Harmeet Kaur3
1Department of Imaging Physics, The University of Texas M.D. Anderson Cancer Center, Houston, TX, United States, 2MR Applications and Workflow, GE Healthcare, Waukesha, WI, United States, 3Department of Radiology, The University of Texas M.D. Anderson Cancer Center, Houston, TX, United States, 4Department of Surgical Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, TX, United States, 5Department of Pathology, The University of Texas M.D. Anderson Cancer Center, Houston, TX, United States

Proper evaluation of rectal cancer with MR imaging requires high resolution imaging of the rectal wall. The image quality demands are difficult to achieve due to the increasing risk of peristaltic motion with longer scan times. In this work, we apply a novel deep learning based reconstruction (DL recon) method to an accelerated sequence using reduced averages and increased acceleration. Radiologist scores indicate that the combined method provides superior SNR and definition with less motion degradation when compared to the routine sequence with conventional reconstruction. Thus improved motion robustness can be gained from applying DL Recon to an accelerated sequence.

This abstract and the presentation materials are available to 2020 meeting attendees and eLibrary customers only; a login is required.

Join Here