This is the code repository for "HARP: Harmonization and Adaptive Refinement of Pseudo-Labels for Cross-Domain Medical Image Segmentation", submitted to MICCAI 2025.

Included Code

1. CDH.py

This code implements image transformation functionality using Singular Value Decomposition (SVD). It processes input images through decomposition, singular value fusion, and reconstruction, supporting both multi-channel and single-channel image processing.

2. APS.py

This code handles pseudo-label selection for semi-supervised learning. It generates pseudo-labels using two models, calculates confidence scores, filters low-confidence labels, and selects samples for annotation based on a specified budget.