.png){kind=icon}
.png){kind=icon}
Return
Structure-Aware Kolmogorov-Arnold Network with Anatomical Priors for Cardiac Segmentation
.png){kind=icon}
PRE
AI
S
S
X
DOI：10.1016/j.asoc.2026.116137.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A novel Structure-Preserving Kolmogorov-Arnold Network (SKANet) is proposed. • Dynamic Group Rational KAN (DGRKAN) enhances non-linear feature refinement. • Multi-scale KA Attention (MSKAM) balances context aggregation and transparency. • Anatomical priors are integrated via Entity Region Consistency Constraint Loss. • Superior performance on ACDC, MSCMRSeg, and M&Ms datasets is demonstrated
Journal
IF:
6.6
Papers:
1.4W
Citations:
4.8W
Organization
Cited Papers