HTC-SGA Former is a hybrid CNN-Transformer model with MS-GLWA, SGFA, and BWACL that outperforms 14 prior methods on private coronary DSA datasets using only 0.81M parameters.
PeerJ Computer Science 8, e993
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
HTC-SGA Former: A Hybrid Transformer-CNN Network with Self-Guided Attention and a New Boundary-Weighted Adaptive Loss for Coronary DSA Vessel Segmentation
HTC-SGA Former is a hybrid CNN-Transformer model with MS-GLWA, SGFA, and BWACL that outperforms 14 prior methods on private coronary DSA datasets using only 0.81M parameters.