Amplitude-Guided Deep Reinforcement Learning for Semi-Supervised Layer Segmentation

The study introduces an Amplitude-guided Deep Reinforcement Learning (ADRL) framework for semi-supervised segmentation of scalp tissue layers, crucial for understanding androgenetic alopecia (AGA). The method integrates a novel data augmentation strategy using Fourier transform, a Phase Alignment (PHA) strategy to reduce noise impact, and a Cross-Power Spectrum Correlation (CPSC) module to improve segmentation accuracy. Experiments on scalp HR-MR and retinal OCT datasets show that this approach significantly outperforms existing methods, addressing challenges like limited labeled data and image quality issues.