Meeting Banner
Abstract #3382

A cross-modality deep learning model for esophageal cancer segmentation and quantitation on 18F-FDG PET/CT and diffusion weighted MRI

Zijian Zhou1, Bikash Panthi1, David E. Rauch1, Jong Bum Son1, Carol C. Wu1, Steven H. Lin1, Mark D. Pagel1, and Jingfei Ma1
1The University of Texas MD Anderson Cancer Center, Houston, TX, United States


We applied a deep learning (DL) model developed for 18F-FDG PET/CT of mantle cell lymphoma to esophageal cancers on 18F-FDG PET/CT and diffusion-weighted MRI. We compared the performance of the DL-based segmentation with the manual segmentation on PET and evaluated the quantitation on both PET and apparent diffusion coefficient (ADC). The model achieved promising results of detecting and segmenting esophageal cancers, and the DL-based imaging metrics were consistent with the reference standards.

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

Join Here