Toward Unpaired Multi-modal Medical Image Segmentation via Learning Structured Semantic Consistency

被引:0
|
作者
Yang, Jie [1 ]
Zhu, Ye [1 ]
Wang, Chaoqun [1 ]
Li, Zhen [1 ]
Zhang, Ruimao [1 ]
机构
[1] Chinese Univ Hong Kong, Shenzhen Res Inst Big Data, Shenzhen, Peoples R China
基金
国家重点研发计划; 中国国家自然科学基金;
关键词
Unpaired multi-modal learning; Structured semantic consistency learning; Medical image segmentation;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Integrating multi-modal data to promote medical image analysis has recently gained great attention. This paper presents a novel scheme to learn the mutual benefits of different modalities to achieve better segmentation results for unpaired multi-modal medical images. Our approach tackles two critical issues of this task from a practical perspective: (1) how to effectively learn the semantic consistencies of various modalities (e.g., CT and MRI), and (2) how to leverage the above consistencies to regularize the network learning while preserving its simplicity. To address (1), we leverage a carefully designed External Attention Module (EAM) to align semantic class representations and their correlations of different modalities. To solve (2), the proposed EAM is designed as an external plug-and-play one, which can be discarded once the model is optimized. We have demonstrated the effectiveness of the proposed method on two medical image segmentation scenarios: (1) cardiac structure segmentation, and (2) abdominal multi-organ segmentation. Extensive results show that the proposed method outperforms its counterparts by a wide margin.
引用
收藏
页码:1602 / 1622
页数:21
相关论文
共 50 条
  • [21] Multi-Modal Medical Image Matching Based on Multi-Task Learning and Semantic-Enhanced Cross-Modal Retrieval
    Zhang, Yilin
    TRAITEMENT DU SIGNAL, 2023, 40 (05) : 2041 - 2049
  • [22] Pseudo Multi-Modal Approach to LiDAR Semantic Segmentation
    Kim, Kyungmin
    SENSORS, 2024, 24 (23)
  • [23] MULTI-MODAL SEMANTIC MESH SEGMENTATION IN URBAN SCENES
    Laupheimer, Dominik
    Haala, Norbert
    XXIV ISPRS CONGRESS IMAGING TODAY, FORESEEING TOMORROW, COMMISSION II, 2022, 5-2 : 267 - 274
  • [24] A nested self-supervised learning framework for 3-D semantic segmentation-driven multi-modal medical image fusion
    Zhang, Ying
    Nie, Rencan
    Cao, Jinde
    Ma, Chaozhen
    Tan, Mingchuan
    BIOMEDICAL SIGNAL PROCESSING AND CONTROL, 2025, 105
  • [25] Interpretable medical image Visual Question Answering via multi-modal relationship graph learning
    Hu, Xinyue
    Gu, Lin
    Kobayashi, Kazuma
    Liu, Liangchen
    Zhang, Mengliang
    Harada, Tatsuya
    Summers, Ronald M.
    Zhu, Yingying
    MEDICAL IMAGE ANALYSIS, 2024, 97
  • [26] Semi-supervised multi-modal medical image segmentation with unified translation
    Sun H.
    Wei J.
    Yuan W.
    Li R.
    Computers in Biology and Medicine, 2024, 176
  • [27] PEMMA: Parameter-Efficient Multi-Modal Adaptation for Medical Image Segmentation
    Saadi, Nada
    Saeed, Numan
    Yaqub, Mohammad
    Nandakumar, Karthik
    MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2024, PT XII, 2024, 15012 : 262 - 271
  • [28] A novel method of medical image segmentation based on the multi-modal function optimization
    Liu, Z. (lxxc1016@gmail.com), 1600, Binary Information Press, Flat F 8th Floor, Block 3, Tanner Garden, 18 Tanner Road, Hong Kong (11):
  • [29] TranSiam: Aggregating multi-modal visual features with locality for medical image segmentation
    Li, Xuejian
    Ma, Shiqiang
    Xu, Junhai
    Tang, Jijun
    He, Shengfeng
    Guo, Fei
    EXPERT SYSTEMS WITH APPLICATIONS, 2024, 237
  • [30] CEFusion: Multi-Modal medical image fusion via cross encoder
    Zhu, Ya
    Wang, Xue
    Chen, Luping
    Nie, Rencan
    IET Image Processing, 2023, 16 (12) : 3177 - 3189