Medical lesion segmentation by combining multimodal images with modality weighted UNet.

Journal: Medical physics
Published Date: Apr 7, 2022

Abstract

PURPOSE: Automatic segmentation of medical lesions is a prerequisite for efficient clinic analysis. Segmentation algorithms for multimodal medical images have received much attention in recent years. Different strategies for multimodal combination (or fusion), such as probability theory, fuzzy models, belief functions, and deep neural networks, have also been developed. In this paper, we propose the modality weighted UNet (MW-UNet) and attention-based fusion method to combine multimodal images for medical lesion segmentation.

Authors

  • Xiner Zhu
    College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China.
  • Yichao Wu
    College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China.
  • Haoji Hu
    College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, China.
  • Xianwei Zhuang
    College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China.
  • Jincao Yao
    Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Zhejiang Cancer Hospital, Hangzhou Institute of Medicine (HIM), Chinese Academy of Sciences, Hangzhou, China.
  • Di Ou
    Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Zhejiang Cancer Hospital, Hangzhou Institute of Medicine (HIM), Chinese Academy of Sciences, Hangzhou, China.
  • Wei Li
    Department of Nephrology, The Second Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi, China.
  • Mei Song
    School of Electronic Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China.
  • Na Feng
    Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Zhejiang Cancer Hospital, Hangzhou Institute of Medicine (HIM), Chinese Academy of Sciences, Hangzhou, China.
  • Dong Xu
    Department of Electrical Engineering and Computer Science, University of Missouri, Columbia, MO, USA.

Keywords

External Resources

Popular Topics
Recent Journals