US2Mask: Image-to-mask generation learning via a conditional GAN for cardiac ultrasound image segmentation

Gang WANG, Mingliang ZHOU*, Xin NING, Prayag TIWARI, Haobo ZHU, Guang YANG, Choon Hwai YAP

*Corresponding author for this work

Research output: Journal PublicationsJournal Article (refereed)peer-review

Abstract

Cardiac ultrasound (US) image segmentation is vital for evaluating clinical indices, but it often demands a large dataset and expert annotations, resulting in high costs for deep learning algorithms. To address this, our study presents a framework utilizing artificial intelligence generation technology to produce multi-class RGB masks for cardiac US image segmentation. The proposed approach directly performs semantic segmentation of the heart's main structures in US images from various scanning modes. Additionally, we introduce a novel learning approach based on conditional generative adversarial networks (CGAN) for cardiac US image segmentation, incorporating a conditional input and paired RGB masks. Experimental results from three cardiac US image datasets with diverse scan modes demonstrate that our approach outperforms several state-of-the-art models, showcasing improvements in five commonly used segmentation metrics, with lower noise sensitivity. Source code is available at https://github.com/energy588/US2mask.

Original languageEnglish
Article number108282
Number of pages13
JournalComputers in Biology and Medicine
Volume172
DOIs
Publication statusPublished - Apr 2024
Externally publishedYes

Bibliographical note

Acknowledgments:
This work was supported in part by the National Natural Science Foundation of China under Grant 62176027; in part by Ningbo Natural Science Foundation under Grant 2023J280, in part by Ningbo Key R&D Program under Grant 2023Z231, in part by Zhejiang Province Postdoctoral Research Funding Project under Grant ZJ2023008, in part by China Postdoctoral Science Foundation under Grant No. 2023M740741; in part by the Key Projects of Basic Strengthening Plan under Grant 2022-JCJQ-ZD-018-11; in part by the Chongqing Talent under Grant cstc2024ycjh-bgzxm0082; in part by the Joint Equipment Pre Research and Key Fund Project of the Ministry of Education under Grant 8091B012207; in part by the Natural Science Foundation of Chongqing, China under Grant cstc2020jcyj-zdxmX0014; in part by the Human Resources and Social Security Bureau Project of Chongqing under Grant cx2020073; and in part by the Guangdong Oppo Mobile Telecommunications Corporation Ltd., under Grant H20221694.

Keywords

  • Artificial intelligence generation
  • Cardiac ultrasound image
  • Image segmentation
  • Mask learning

Fingerprint

Dive into the research topics of 'US2Mask: Image-to-mask generation learning via a conditional GAN for cardiac ultrasound image segmentation'. Together they form a unique fingerprint.

Cite this