Segmentation Loss Odyssey
This work addresses the problem of fragmented loss function studies for researchers in medical image segmentation, though it is incremental as it organizes existing methods rather than introducing new ones.
The authors tackled the lack of systematic comparison in medical image segmentation loss functions by proposing a taxonomy to categorize existing ones into four groups, revealing fundamental similarities and exploring relationships between region-based and boundary-based losses, with PyTorch implementations made publicly available.
Loss functions are one of the crucial ingredients in deep learning-based medical image segmentation methods. Many loss functions have been proposed in existing literature, but are studied separately or only investigated with few other losses. In this paper, we present a systematic taxonomy to sort existing loss functions into four meaningful categories. This helps to reveal links and fundamental similarities between them. Moreover, we explore the relationship between the traditional region-based and the more recent boundary-based loss functions. The PyTorch implementations of these loss functions are publicly available at \url{https://github.com/JunMa11/SegLoss}.