.png){kind=icon}
.png){kind=icon}
Return
DiSCNet: Directional Split Convolution for compute-efficient brain tumor diagnosis
.png){kind=icon}
PRE
AI
DOI：10.1016/j.compbiolchem.2026.109066.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• DiSCNet achieves 99.22% accuracy for four-class brain tumor MRI classification. • Directional Split Convolution captures local and anisotropic tumor cues. • GRN and ECA improve robustness to scanner and protocol variability. • The model outperforms 71 CNN, ViT, and hybrid baselines. • DiSCNet delivers top performance with only 2.78 M parameters.
Keywords:
DiSCNet
Directional Split Convolution
brain tumor classification
MRI
compute-efficient
Journal
C
IF:
3.1
Papers:
391
Citations:
0