핵심 개념
提案されたDLKおよびDFFモジュールを組み込んだD-Netは、多様な形状とサイズの臓器からマルチスケール特徴を効果的に捉え、グローバルコンテキスト情報を適応的に活用することで、3D体積医用画像セグメンテーションにおいて優れた性能を発揮します。
통계
Hierarchical transformers have achieved significant success in medical image segmentation.
CNNs incorporated with large convolutional kernels remain constrained in adaptively capturing multi-scale features.
Extensive experimental results demonstrate that D-Net outperforms other state-of-the-art models.
인용구
"DLK module employs multiple large convolutional kernels to capture multi-scale features."
"DFF is designed to adaptively fuse multi-scale local features based on global information."