On the influence of Dice loss function in multi-class organ segmentation of abdominal CT using 3D fully convolutional networks

Deep learning-based methods achieved impressive results for the segmentation medical images. With the development of 3D fully convolutional networks(FCNs), it has become feasible to produce improved results for multi-organ of 3D computed tomography (CT) images. The results of multi-organ using deep learning-based methods not only depend on the choice of architecture, but also strongly rely on the choice of loss function. this paper, we present a discussion on the influence of Dice-based loss for multi-class organ segmentation using a dataset of abdominal CT. We investigated three different types of weighting the Dice loss based on class label frequencies (uniform, simple and square) and their influence on segmentation accuracies. Furthermore, we compared influence of different initial learning rates. We achieved average Dice of 81.3%, 59.5% and 31.7% for uniform, simple and square types of when the learning rate is 0.001, and 78.2%, 81.0% and 58.5% for each when the learning rate is 0.01. Our experiments indicated a strong between class balancing weights and initial learning rate in.