Background <p>Semi-Supervised Learning reduces the annotation burden for medical image segmentation but currently suffers from spatial context overfitting and confirmation bias due to noisy pseudo-labels. This study aims to propose a robust framework to overcome these limitations.</p> Methods <p>We introduce a novel framework termed Spatially Decoupled Reliable Mutual Learning (SDRML). To address context overfitting, we propose a Spatial Decoupling strategy that utilizes translation consistency, compelling the model to focus on intrinsic anatomical features rather than fixed background contexts. To mitigate confirmation bias, we design a Reliable Mutual Learning mechanism incorporating a Confident Regional Cross-entropy loss. This loss dynamically filters low-confidence predictions, ensuring only reliable pseudo-labels guide the tri-model co-training process.</p> Results <p>Extensive experiments were conducted on the ACDC (2D MRI), Left Atrium (3D MRI), and Pancreas-CT datasets. SDRML significantly outperforms state-of-the-art methods across all benchmarks. Notably, it demonstrates superior robustness and segmentation accuracy in data-scarce scenarios, such as regimes with only 10% labeled data.</p> Conclusions <p>SDRML effectively resolves spatial dependency and noise accumulation issues in SSL. By leveraging spatial decoupling and reliable noise filtering, it provides a highly effective solution for medical image segmentation with limited annotations.</p> Trial registration <p>Not applicable.</p>