Abstract #1729

Proton density fat fraction results derived from deep learning auto-segmentation correlate strongly with results obtained by manual analysis

Ashley L. Louie¹, Kang Wang¹, Timoteo Delgado¹, Michael S. Middleton¹, Gavin Hamilton¹, Tanya Wolfson², Robert P. Myers³, C. Stephen Djedjos³, Rohit Loomba⁴, and Claude B. Sirlin¹

¹Liver Imaging Group, Department of Radiology, University of California San Diego, La Jolla, CA, United States, ²Computational and Applied Statistics Laboratory (CASL), SDSC, University of California San Diego, La Jolla, CA, United States, ³Gilead Sciences, Inc., Foster City, CA, United States, ⁴NAFLD Research Center, Division of Gastroenterology, University of California San Diego, La Jolla, CA, United States

A widely-accepted method to estimate hepatic proton-density fat fraction (PDFF) is by averaging values derived from manually drawn regions-of-interest (ROIs) in the nine Couinaud segments. An automated deep-learning-based segmentation tool has been developed to potentially replace this labor-intensive and technically-challenging method. The purpose of this study was to compare whole-liver PDFF values obtained using this auto-segmentation tool to results obtained using manual analysis for a longitudinal multi-center clinical trial of 72 patients with nonalcoholic steatohepatitis. We found that PDFF values estimated using the auto-segmentation tool were in near agreement with values derived by manually drawing ROIs.

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

Join Here