GNAS-U2Net: A New Optic Cup and Optic Disc Segmentation Architecture With Genetic Neural Architecture Search

• Published in: IEEE signal processing letters Vol. 29; pp. 697 - 701
• Main Authors: Sun, Jun-Ding, Yao, Chong, Liu, Jie, Liu, Weifan, Yu, Ze-Kuan
• Format: Journal Article
• Language: English
• Published: New York IEEE 2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Adaptive optics; Artificial intelligence; Biomedical optical imaging; Computer architecture; convolutional neural network; Datasets; Image segmentation; Medical imaging; Microprocessors; neural architecture search; Optical imaging; Searching; Task analysis
• ISSN: 1070-9908; 1558-2361
• DOI: 10.1109/LSP.2022.3151549

Neural architecture search (NAS) has made incredible progress in medical image segmentation tasks, due to its automatic design of the model. However, the search spaces studied in many existing studies are based on U-Net and its variants, which limits the potential of neural architecture search in modeling better architectures. In this study, we propose a new NAS architecture named GNAS-U 2 Net for the joint segmentation of optic cup and optic disc. This architecture is the first application of NAS in a two-level nested U-shaped structure. The best performance achieved by the joint segmentation model designed by NAS on the REFUGE dataset has an average DICE of 92.88%. Compared to U 2 -Net and other related work, the model has better performance and uses only 34.79M parameters. We then verify the generalization of the model on two datasets, namely the Drishti-GS dataset and the GAMMA dataset, for which we obtain an average DICE of 92.32% and 92.11% respectively.