Robust T-Loss for Medical Image Segmentation
This addresses the challenge of label noise in medical image segmentation, which is common due to human error, offering a practical solution for improving model robustness in this domain.
The paper tackles the problem of medical image segmentation with noisy labels by introducing the T-Loss, a robust loss function based on the Student-t distribution that adaptively handles outliers. Results show it outperforms traditional loss functions, achieving higher dice scores on skin lesion and lung segmentation datasets.
This paper presents a new robust loss function, the T-Loss, for medical image segmentation. The proposed loss is based on the negative log-likelihood of the Student-t distribution and can effectively handle outliers in the data by controlling its sensitivity with a single parameter. This parameter is updated during the backpropagation process, eliminating the need for additional computation or prior information about the level and spread of noisy labels. Our experiments show that the T-Loss outperforms traditional loss functions in terms of dice scores on two public medical datasets for skin lesion and lung segmentation. We also demonstrate the ability of T-Loss to handle different types of simulated label noise, resembling human error. Our results provide strong evidence that the T-Loss is a promising alternative for medical image segmentation where high levels of noise or outliers in the dataset are a typical phenomenon in practice. The project website can be found at https://robust-tloss.github.io