Research Article
Retinal Vessel Segmentation Using Multi-scale Generative Adversarial Network with Class Activation Mapping
@INPROCEEDINGS{10.1007/978-3-031-06368-8_7, author={Minqiang Yang and Yinru Ye and Kai Ye and Xiping Hu and Bin Hu}, title={Retinal Vessel Segmentation Using Multi-scale Generative Adversarial Network with Class Activation Mapping}, proceedings={Wireless Mobile Communication and Healthcare. 10th EAI International Conference, MobiHealth 2021, Virtual Event, November 13--14, 2021, Proceedings}, proceedings_a={MOBIHEALTH}, year={2022}, month={6}, keywords={Retinal vessel segmentation Multi-scale generative adversarial network Class activation mapping Data augmentation}, doi={10.1007/978-3-031-06368-8_7} }- Minqiang Yang
Yinru Ye
Kai Ye
Xiping Hu
Bin Hu
Year: 2022
Retinal Vessel Segmentation Using Multi-scale Generative Adversarial Network with Class Activation Mapping
MOBIHEALTH
Springer
DOI: 10.1007/978-3-031-06368-8_7
Abstract
Retinal vessel segmentation plays a significant role in the accurate diagnosis of retinal diseases. However, existing methods commonly omit micro-vessels in retinal images and generate some false-positive vessels. To alleviate this issue, we propose a multi-scale generative adversarial network with class activation mapping to achieve efficient segmentation. For the problem of small amount of data, we introduce a novel data augmentation method, which can generate multiple samples by cutting pixels from other samples. This method increases the diversity of samples and improve the robustness of the model. We compare our method with previous models with several metrics, and experiments show the superiority and effectiveness of our model.
