C2FNAS: Coarse-to-Fine Neural Architecture Search for 3D Medical Image Segmentation
Qihang Yu, Dong Yang, Holger Roth, Yutong Bai, Yixiao Zhang, Alan L. Yuille, Daguang Xu
3D convolution neural networks (CNN) have been proved very successful in parsing organs or tumours in 3D medical images, but it remains sophisticated and time-consuming to choose or design proper 3D networks given different task contexts. Recently, Neural Architecture Search (NAS) is proposed to solve this problem by searching for the best network architecture automatically. However, the inconsistency between search stage and deployment stage often exists in NAS algorithms due to memory constraints and large search space, which could become more serious when applying NAS to some memory and time-consuming tasks, such as 3D medical image segmentation. In this paper, we propose a coarse-to-fine neural architecture search (C2FNAS) to automatically search a 3D segmentation network from scratch without inconsistency on network size or input size. Specifically, we divide the search procedure into two stages: 1) the coarse stage, where we search the macro-level topology of the network, i.e. how each convolution module is connected to other modules; 2) the fine stage, where we search at micro-level for operations in each cell based on previous searched macro-level topology. The coarse-to-fine manner divides the search procedure into two consecutive stages and meanwhile resolves the inconsistency. We evaluate our method on 10 public datasets from Medical Segmentation Decalthon (MSD) challenge, and achieve state-of-the-art performance with the network searched using one dataset, which demonstrates the effectiveness and generalization of our searched models.
@INPROCEEDINGS{9156654, author={Q. {Yu} and D. {Yang} and H. {Roth} and Y. {Bai} and Y. {Zhang} and A. L. {Yuille} and D. {Xu}}, booktitle={2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, title={C2FNAS: Coarse-to-Fine Neural Architecture Search for 3D Medical Image Segmentation}, year={2020}, volume={}, number={}, pages={4125-4134}, doi={10.1109/CVPR42600.2020.00418}}