LesionMix: A Lesion-Level Data Augmentation Method for Medical Image Segmentation

被引:1
|
作者
Basaran, Berke Doga [1 ,2 ]
Zhang, Weitong [1 ]
Qiao, Mengyun [2 ,3 ]
Kainz, Bernhard [1 ,4 ]
Matthews, Paul M. [3 ,5 ]
Bai, Wenjia [1 ,2 ,3 ]
机构
[1] Imperial Coll London, Dept Comp, London, England
[2] Imperial Coll London, Data Sci Inst, London, England
[3] Imperial Coll London, Dept Brain Sci, London, England
[4] Friedrich Alexander Univ Erlangen Nurnberg, Erlangen, DE, Germany
[5] Imperial Coll London, UK Dementia Res Inst, London, England
关键词
Data augmentation; Lesion populating; Lesion inpainting; Image synthesis; Lesion image segmentation;
D O I
10.1007/978-3-031-58171-7_8
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Data augmentation has become a de facto component of deep learning-based medical image segmentation methods. Most data augmentation techniques used in medical imaging focus on spatial and intensity transformations to improve the diversity of training images. They are often designed at the image level, augmenting the full image, and do not pay attention to specific abnormalities within the image. Here, we present LesionMix, a novel and simple lesion-aware data augmentation method. It performs augmentation at the lesion level, increasing the diversity of lesion shape, location, intensity and load distribution, and allowing both lesion populating and inpainting. Experiments on different modalities and different lesion datasets, including four brain MR lesion datasets and one liver CT lesion dataset, demonstrate that LesionMix achieves promising performance in lesion image segmentation, outperforming several recent Mix-based data augmentation methods. The code will be released at https://github.com/dogabasaran/lesionmix.
引用
收藏
页码:73 / 83
页数:11
相关论文
共 50 条
  • [31] Cross-set data augmentation for semi-supervised medical image segmentation
    Wu, Qianhao
    Jiang, Xixi
    Zhang, Dong
    Feng, Yifei
    Tanga, Jinhui
    IMAGE AND VISION COMPUTING, 2025, 154
  • [32] Evaluating Medical Image Segmentation Models Using Augmentation
    Sayed, Mattin
    Saba-Sadiya, Sari
    Wichtlhuber, Benedikt
    Dietz, Julia
    Neitzel, Matthias
    Keller, Leopold
    Roig, Gemma
    Bucher, Andreas M.
    TOMOGRAPHY, 2024, 10 (12) : 2128 - 2143
  • [33] A software framework for preprocessing and level set segmentation of medical image data
    Fritscher, KD
    Schubert, R
    MEDICAL IMAGING 2005: IMAGE PROCESSING, PT 1-3, 2005, 5747 : 1742 - 1752
  • [34] PIXEL LEVEL DATA AUGMENTATION FOR SEMANTIC IMAGE SEGMENTATION USING GENERATIVE ADVERSARIAL NETWORKS
    Liu, Shuangting
    Zhang, Jiaqi
    Chen, Yuxin
    Liu, Yifan
    Qin, Zengchang
    Wan, Tao
    2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2019, : 1902 - 1906
  • [35] Registration Based Data Augmentation for Multiple Sclerosis Lesion Segmentation
    Abolvardi, Ava Assadi
    Hamey, Len
    Ho-Shon, Kevin
    2019 DIGITAL IMAGE COMPUTING: TECHNIQUES AND APPLICATIONS (DICTA), 2019, : 408 - 412
  • [36] Modified fast marching and level set method for medical image segmentation
    Zhu, Fuping
    Tian, Jie
    2003, IOS Press (11)
  • [37] New Medical Image Sequences Segmentation Based on Level Set Method
    Qin, Xujia
    Zhang, Suqiong
    PROCEEDINGS OF 2009 INTERNATIONAL CONFERENCE ON IMAGE ANALYSIS AND SIGNAL PROCESSING, 2009, : 22 - 27
  • [38] A new stopping force to level set method for medical image segmentation
    Shreedhara, KS
    Kumar, AA
    Proceedings of the IEEE INDICON 2004, 2004, : 191 - 194
  • [39] Medical Image Segmentation Based on Wavelet Transformation and Level set method
    Larbi, Messaouda
    Rouini, Abdelghani
    Messali, Zoubeida
    Larbi, Samira
    2019 3RD INTERNATIONAL CONFERENCE ON APPLIED AUTOMATION AND INDUSTRIAL DIAGNOSTICS (ICAAID 2019), 2019,
  • [40] A level set method with global and local information for medical image segmentation
    Zhu, Q. (qingzhu@yzu.edu.com), 1600, Central South University of Technology (44):