The usability of generative adversarial networks for automatic segmentation of lung nodules in CT-images

Lung cancer is a highly prevailing disease and early detection and treatment are crucial to increase the likelihood of survival. In many lung cancer procedures, the segmentation of lung nodules in CT images is an essential step. However, manual annotation of the nodules is a difficult and time-consuming task, which relies heavily on the experience of a radiologist. To assist radiologists in this process, computer-assisted segmentation systems could be a promising tool.

The aim of this study was to explore to what extent generative adversarial networks (GANs) can be used to automatically segment lung nodules from entire 2D CT images. The overall network implemented in this research followed the structure of a conditional image-to-image translation GAN, in which a U-Net was used as the generator network. Three modules were added to the architecture to address three commonly occurring segmentation challenges: (i) variability in lung nodule appearance, (ii) class imbalance, and (iii) GAN training instability. The added value of each module to the generator network was evaluated in an ablation study. Additionally, the usefulness of GANs compared to the state-of-the-art segmentation network, the U-Net, was explored.

To join the presentation via Microsoft Teams click here