Enhancing source-free domain adaptation in Medical Image Segmentation via regulated model self-training

Tianwei Zhang, Kang Li, Shi Gu, Pheng-Ann Heng

Abstract: Highlights•Novel source-free domain adaptation (SFDA) approach for a broad range of applications•Investigating SFDA from a new perspective, i.e., knowledge transfer•Seamlessly work with prior mainstream pseudo-label denoising methods for more gains•State-of-the-art performance under major and minor domain shifts