J Shanghai Jiaotong Univ Sci ›› 2021, Vol. 26 ›› Issue (1): 93-102.doi: 10.1007/s12204-021-2264-x

• Medicine-Engineering Interdisciplinary Research • Previous Articles     Next Articles

Rethinking the Dice Loss for Deep Learning Lesion Segmentation in Medical Images

Rethinking the Dice Loss for Deep Learning Lesion Segmentation in Medical Images

ZHANG Yue (张月), LIU Shijie (刘世界), LI Chunlai (李春来), WANG Jianyu (王建宇)   

  1. (1. Key Laboratory of Space Active Opto-Electronics Technology, Shanghai Institute of Technical Physics, Chinese
    Academy of Sciences, Shanghai 200083, China; 2. University of Chinese Academy of Sciences, Beijing 100049, China;
    3. Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences, Hangzhou 310024, China)
  2. (1. Key Laboratory of Space Active Opto-Electronics Technology, Shanghai Institute of Technical Physics, Chinese
    Academy of Sciences, Shanghai 200083, China; 2. University of Chinese Academy of Sciences, Beijing 100049, China;
    3. Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences, Hangzhou 310024, China)
  • Online:2021-02-28 Published:2021-01-19
  • Contact: WANG Jianyu (王建宇) E-mail: jywang@mail.sitp.ac.cn

Abstract: Deep learning is widely used for lesion segmentation in medical images due to its breakthrough performance. Loss functions are critical in a deep learning pipeline, and they play important roles in segmenting performance. Dice loss is the most commonly used loss function in medical image segmentation, but it also has some disadvantages. In this paper, we discuss the advantages and disadvantages of the Dice loss function, and group the extensions of the Dice loss according to its improved purpose. The performances of some extensions are compared according to core references. Because different loss functions have different performances in different tasks, automatic loss function selection will be the potential direction in the future.


Key words: Dice loss| deep learning| medical image| lesion segmentation

摘要: Deep learning is widely used for lesion segmentation in medical images due to its breakthrough performance. Loss functions are critical in a deep learning pipeline, and they play important roles in segmenting performance. Dice loss is the most commonly used loss function in medical image segmentation, but it also has some disadvantages. In this paper, we discuss the advantages and disadvantages of the Dice loss function, and group the extensions of the Dice loss according to its improved purpose. The performances of some extensions are compared according to core references. Because different loss functions have different performances in different tasks, automatic loss function selection will be the potential direction in the future.


关键词: Dice loss| deep learning| medical image| lesion segmentation

CLC Number: