오유진 (KAIST) | 한빛사논문 > 한빛사

한빛사논문

조회533

오유진Yujin Oh

KAIST

Med. Image Anal. , Nov 11 2023, 91 S1361-8415(23)00282-7 | https://doi.org/10.1016/j.media.2023.103022 C-DARL: Contrastive diffusion adversarial representation learning for label-free blood vessel segmentation

Boah Kim ^a,1, Yujin Oh ^b,1, Bradford J. Wood ^a, Ronald M. Summers ^a, Jong Chul Ye ^b

^aRadiology and Imaging Sciences, National Institutes of Health Clinical Center, Bethesda, MD, USA

^bKim Jaechul Graduate School of AI, Korea Advanced Institute of Science & Technology (KAIST), Daejeon, Republic of Korea

¹Co-first authors.

²This paper extends the work (Kim et al., 2023) presented at the Eleventh International Conference on Learning Representations (ICLR) 2023.

Corresponding authors: Ronald M. Summers, Jong Chul Ye

Abstract

Blood vessel segmentation in medical imaging is one of the essential steps for vascular disease diagnosis and interventional planning in a broad spectrum of clinical scenarios in image-based medicine and interventional medicine. Unfortunately, manual annotation of the vessel masks is challenging and resource-intensive due to subtle branches and complex structures. To overcome this issue, this paper presents a self-supervised vessel segmentation method, dubbed the contrastive diffusion adversarial representation learning (C-DARL) model. Our model is composed of a diffusion module and a generation module that learns the distribution of multi-domain blood vessel data by generating synthetic vessel images from diffusion latent. Moreover, we employ contrastive learning through a mask-based contrastive loss so that the model can learn more realistic vessel representations. To validate the efficacy, C-DARL is trained using various vessel datasets, including coronary angiograms, abdominal digital subtraction angiograms, and retinal imaging. Experimental results confirm that our model achieves performance improvement over baseline methods with noise robustness, suggesting the effectiveness of C-DARL for vessel segmentation.Our source code is available at https://github.com/boahK/MEDIA_CDARL. ².

논문정보

형식Research article
게재일2023년 11월 (BRIC 등록일 2023-11-21)
연구진국내(교신)+국외 연구진
분야 바이오·의료융합 > 의료기기

댓글 작성 로그인

CV 열람하기

연락처 보기