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

被引：42

作者：

Zhang Y. ^{[1
,2
]}

Liu S. ^{[1
,2
,3
]}

Li C. ^{[1
]}

Wang J. ^{[1
,2
,3
]}

机构：

[1] Key Laboratory of Space Active Opto-Electronics Technology, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, Shanghai

[2] University of Chinese Academy of Sciences, Beijing

[3] Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences, Hangzhou

来源：

Journal of Shanghai Jiaotong University (Science) | 2021年 / 26卷 / 01期

关键词：

A; deep learning; Dice loss; lesion segmentation; medical image; R; 857.3; TP; 391.4;

D O I：

10.1007/s12204-021-2264-x

中图分类号：

学科分类号：

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. © 2020, Shanghai Jiao Tong University and Springer-Verlag GmbH Germany, part of Springer Nature.

引用

页码：93 / 102

页数：9

共 54 条

[1]

Garcea-Lorenzo D., Francis S., Narayanan S., Et al., Review of automatic segmentation methods of multiple sclerosis white matter lesions on conventional magnetic resonance imaging [J], Medical Image Analysis, 17, 1, pp. 1-18, (2013)

[2]

Saritha S., Prabha N.A., A comprehensive review: Segmentation of MRI images — brain tumor [J], International Journal of Imaging Systems and Technology, 26, 4, pp. 295-304, (2016)

[3]

Wadhwa A., Bhardwaj A., Verma V.S., A review on brain tumor segmentation of MRI images [J], Magnetic Resonance Imaging, 61, pp. 247-259, (2019)

[4]

Weeda M.M., Brouwer I., de Vos M.L., Et al., Comparing lesion segmentation methods in multiple sclerosis: Input from one manually delineated subject is sufficient for accurate lesion segmentation [J], NeuroImage: Clinical, 24, (2019)

[5]

Bernal J., Kushibar K., Asfaw D.S., Et al., Deep convolutional neural networks for brain image analysis on magnetic resonance imaging: A review [J], Artificial Intelligence in Medicine, 95, pp. 64-81, (2019)

[6]

Kervadec H., Bouchtiba J., Desrosiers C., Et al., Boundary loss for highly unbalanced segmentation [J], Medical Image Analysis, 67, (2021)

[7]

Chen C., Qin C., Qiu H.Q., Et al., Deep learning for cardiac image segmentation: A review [J], Frontiers in Cardiovascular Medicine, 7, (2020)

[8]

Isin A., Direkoglu C., Sah M., Review of MRI-based brain tumor image segmentation using deep learning methods [J], Procedia Computer Science, 102, pp. 317-324, (2016)

[9]

Jaafra Y., Laurent J.L., Deruyver A., Et al., Reinforcement learning for neural architecture search: Areview [J], Image and Vision Computing, 89, pp. 57-66, (2019)

[10]

Makropoulos A., Counsell S.J., Rueckert D., A review on automatic fetal and neonatal brain MRI segmentation [J], NeuroImage, 170, pp. 231-248, (2018)

← 1 2 3 4 5 6 →