Meeting Banner
Abstract #3190

Deep Learning-based Generative Adversarial Registration NETwork (GARNET) for Hepatocellular Carcinoma Segmentation: Multi-center Study

Hang Yu1, Rencheng Zheng1, Weibo Chen2, Ruokun Li3, Huazheng Shi4, Chengyan Wang5, and He Wang1
1Institute of Science and Technology for Brain-Inspired Intelligence, Fudan University, Shanghai, China, 2Philips Healthcare, Shanghai, China, 3Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China, 4Shanghai Universal cloud imaging dignostic center, Shanghai, China, 5Human Phenome Institute, Fudan University, ShangHai, China


This study proposed a deep learning-based generative adversarial registration network (GARNET) for multi-contrast liver image registration and evaluated its value for hepatocellular carcinoma (HCC) segmentation. We used generative adversarial net (GAN) to synthesize images from diffusion-weighted imaging (DWI) to dynamic contrast-enhanced (DCE) and then applied for deformable registration on the synthesized DCE images. A total of 607 cirrhosis patients from 5 centers (401 HCC patients) were included in this study. We compared the proposed method with symmetric image normalization (SyN) registration and VoxelMorph. Experimental results demonstrated that GARNET improved the registration performances significantly and yielded better segmentation of HCC lesions.

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

Join Here