Improving brain tumor segmentation with anatomical prior-informed pre-training

被引:0
|
作者
Wang, Kang [1 ,2 ]
Li, Zeyang [3 ]
Wang, Haoran [1 ,2 ]
Liu, Siyu [1 ,2 ]
Pan, Mingyuan [4 ]
Wang, Manning [1 ,2 ]
Wang, Shuo [1 ,2 ]
Song, Zhijian [1 ,2 ]
机构
[1] Fudan Univ, Digital Med Res Ctr, Sch Basic Med Sci, Shanghai, Peoples R China
[2] Fudan Univ, Shanghai Key Lab Med Image Comp & Comp Assisted In, Shanghai, Peoples R China
[3] Fudan Univ, Zhongshan Hosp, Dept Neurosurg, Shanghai, Peoples R China
[4] Fudan Univ, Huashan Hosp, Radiat Oncol Ctr, Shanghai, Peoples R China
基金
中国国家自然科学基金;
关键词
masked autoencoder; anatomical priors; transformer; brain tumor segmentation; magnetic resonance image; self-supervised learning; OPTIMIZATION; REGISTRATION; ROBUST;
D O I
10.3389/fmed.2023.1211800
中图分类号
R5 [内科学];
学科分类号
1002 ; 100201 ;
摘要
IntroductionPrecise delineation of glioblastoma in multi-parameter magnetic resonance images is pivotal for neurosurgery and subsequent treatment monitoring. Transformer models have shown promise in brain tumor segmentation, but their efficacy heavily depends on a substantial amount of annotated data. To address the scarcity of annotated data and improve model robustness, self-supervised learning methods using masked autoencoders have been devised. Nevertheless, these methods have not incorporated the anatomical priors of brain structures.MethodsThis study proposed an anatomical prior-informed masking strategy to enhance the pre-training of masked autoencoders, which combines data-driven reconstruction with anatomical knowledge. We investigate the likelihood of tumor presence in various brain structures, and this information is then utilized to guide the masking procedure.ResultsCompared with random masking, our method enables the pre-training to concentrate on regions that are more pertinent to downstream segmentation. Experiments conducted on the BraTS21 dataset demonstrate that our proposed method surpasses the performance of state-of-the-art self-supervised learning techniques. It enhances brain tumor segmentation in terms of both accuracy and data efficiency.DiscussionTailored mechanisms designed to extract valuable information from extensive data could enhance computational efficiency and performance, resulting in increased precision. It's still promising to integrate anatomical priors and vision approaches.
引用
收藏
页数:14
相关论文
共 50 条
  • [1] Weakly supervised pre-training for brain tumor segmentation using principal axis measurements of tumor burden
    Mckone, Joshua E.
    Lambrou, Tryphon
    Ye, Xujiong
    Brown, James M.
    FRONTIERS IN COMPUTER SCIENCE, 2024, 6
  • [2] Improving Fractal Pre-training
    Anderson, Connor
    Farrell, Ryan
    2022 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV 2022), 2022, : 2412 - 2421
  • [3] Multi-modal U-Nets with Boundary Loss and Pre-training for Brain Tumor Segmentation
    Lorenzo, Pablo Ribalta
    Marcinkiewicz, Michal
    Nalepa, Jakub
    BRAINLESION: GLIOMA, MULTIPLE SCLEROSIS, STROKE AND TRAUMATIC BRAIN INJURIES (BRAINLES 2019), PT II, 2020, 11993 : 135 - 147
  • [4] Improving work detection by segmentation heuristics pre-training on factory operations video
    Kataoka, Shotaro
    Ito, Tetsuro
    Iwaka, Genki
    Oba, Masashi
    Nonaka, Hirofumi
    PLOS ONE, 2022, 17 (06):
  • [5] Improving fault localization with pre-training
    Zhang, Zhuo
    Li, Ya
    Xue, Jianxin
    Mao, Xiaoguang
    FRONTIERS OF COMPUTER SCIENCE, 2024, 18 (01)
  • [6] LETCP: A Label-Efficient Transformer-Based Contrastive Pre-Training Method for Brain Tumor Segmentation
    Chen, Shoucun
    Zhang, Jing
    Zhang, Tianchi
    APPLIED SCIENCES-BASEL, 2022, 12 (21):
  • [7] Intra-modality masked image modeling: A self-supervised pre-training method for brain tumor segmentation
    Qi, Liangce
    Shi, Weili
    Miao, Yu
    Li, Yonghui
    Feng, Guanyuan
    Jiang, Zhengang
    BIOMEDICAL SIGNAL PROCESSING AND CONTROL, 2024, 95
  • [8] Learning Visual Prior via Generative Pre-Training
    Xie, Jinheng
    Ye, Kai
    Li, Yudong
    Li, Yuexiang
    Lin, Kevin Qinghong
    Zheng, Yefeng
    Shen, Linlin
    Shou, Mike Zheng
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
  • [9] Brain tumor grade classification using multi-step pre-training
    Mehmood, Yasar
    Bajwa, Usama Ijaz
    Anwar, Muhammad Waqas
    INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY, 2024, 34 (02)
  • [10] Pre-training with Meta Learning for Chinese Word Segmentation
    Ke, Zhen
    Shi, Liang
    Sun, Songtao
    Meng, Erli
    Wang, Bin
    Qiu, Xipeng
    2021 CONFERENCE OF THE NORTH AMERICAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS: HUMAN LANGUAGE TECHNOLOGIES (NAACL-HLT 2021), 2021, : 5514 - 5523