Yingzhen Hu

Semi-supervised 3D medical image segmentation reduces the need for dense voxel-level annotations by exploiting unlabeled volumes. Although existing methods such as consistency regularization, pseudo-labeling, and co-training improve prediction-level robustness, they often provide insufficient feature-space organization for anatomically complex structures, especially small organs and ambiguous boundary regions with large intra-class variations.

CONFLICT OF INTEREST STATEMENT The authors declare no conflicts of interest. Supporting Information Filename Description jace70450-sup-0001-SuppMat.docx27.3 MB Supporting Information REFERENCES 1, , , , , , et al. Effect of cathode geometry on the electrochemical performance of flat tubular segmented-in-series(SIS) solid oxide fuel cell. Int J Hydrogen Energy. 2015; 40(18): 6207–6215. https://doi.org/10.1016/j.ijhydene.2015.03.040 2, , , , .