The field of medical image segmentation is witnessing significant developments, with a growing focus on improving the accuracy and robustness of segmentation models. One of the key directions is the integration of multi-modal data, including images and textual information, to enhance the performance of segmentation models. Another area of research is the development of novel frameworks and techniques to alleviate the reliance on large amounts of annotated data, which is often scarce in medical imaging applications. Noteworthy papers in this area include: Cycle Context Verification, which proposes a novel framework for enhancing in-context medical image segmentation by enabling self-verification of predictions and improving contextual alignment. A Multi-Modal Fusion Framework, which develops a novel multi-modal fusion framework for brain tumor segmentation that integrates spatial-language-vision information through bidirectional interactive attention mechanisms. Alleviating Textual Reliance, which proposes ProLearn, the first Prototype-driven Learning framework for language-guided segmentation that fundamentally alleviates textual reliance. Out-of-distribution data supervision, which introduces a data-centric framework to address the issue of unexpected misclassification between foreground and background objects in biomedical segmentation networks. Hybrid Ensemble Approaches, which proposes a novel double ensembling framework for enhanced brain tumor classification. Text-driven Multiplanar Visual Interaction, which proposes a novel text-driven multiplanar visual interaction framework for semi-supervised medical image segmentation.