Semi-supervised medical image segmentation method using multi-scale consistency adversarial learning

doi:10.1016/j.bspc.2025.108250

Semi-supervised medical image segmentation method using multi-scale consistency adversarial learning

2025-07-15

0

PRE

AI

DOI：10.1016/j.bspc.2025.108250

Original

Original request for help

Save

Abstract

En 中文

• A comprehensive perturbation space is constructed to enhance the model’s ability to generalize from limited labeled data. • A strong and weak consistency regularization method is proposed on multiple scales. • An adaptive weighted pyramid consistency loss is proposed to promote consistent results. • Extensive experiments on ACDC and BraTS2019 datasets show superior performance.

Semi-supervised medical image segmentation method using multi-scale consistency adversarial learning

Abstract

Journal

Organization