On the influence of Dice loss function in multi-class organ segmentation of abdominal CT using 3D fully convolutional networks
This work addresses the problem of optimizing segmentation accuracy for medical imaging researchers, but it is incremental as it focuses on tuning existing loss functions and learning rates.
The paper investigated the influence of Dice loss functions with different class weighting schemes and initial learning rates on multi-class organ segmentation in abdominal CT volumes, achieving average Dice scores ranging from 31.7% to 81.3% depending on the configuration.
Deep learning-based methods achieved impressive results for the segmentation of medical images. With the development of 3D fully convolutional networks (FCNs), it has become feasible to produce improved results for multi-organ segmentation of 3D computed tomography (CT) images. The results of multi-organ segmentation using deep learning-based methods not only depend on the choice of networks architecture, but also strongly rely on the choice of loss function. In this paper, we present a discussion on the influence of Dice-based loss functions for multi-class organ segmentation using a dataset of abdominal CT volumes. We investigated three different types of weighting the Dice loss functions based on class label frequencies (uniform, simple and square) and evaluate their influence on segmentation accuracies. Furthermore, we compared the influence of different initial learning rates. We achieved average Dice scores of 81.3%, 59.5% and 31.7% for uniform, simple and square types of weighting when the learning rate is 0.001, and 78.2%, 81.0% and 58.5% for each weighting when the learning rate is 0.01. Our experiments indicated a strong relationship between class balancing weights and initial learning rate in training.