Abstract: The scarcity of labeled data is a challenging problem that affects the segmentation accuracy of medical images. Aiming to solve this problem, we propose a semi-supervised learning method based on fourier transform and consistent constraint, In the case of a small amount of annotated data, the output of unannotated data via Fourier transform interpolation and model segmentation is spa-tially consistent with the output of reverse operation, and the consistency regularization constraint for unannotated data is con-structed to improve the model performance of fully supervised learning. The experimental results on public datasets ACDC, Syn-apse and CTLN show that the proposed algorithm is superior to baseline methods and can be integrated with existing SOTA semi-supervised medical image segmentation methods to improve their segmentation performances.

Key words: medical image segmentation, semi-supervised learning, Fourier transformation, consistency regularization constraint