Lite-MixedNet: Lightweight and efficient hybrid network for medical image segmentation

Open Webpage

Zitong Ren, Yongming Li, Liejun Wang, Lianghui Xu

Published: 2025, Last Modified: 23 Jan 2026Pattern Recognit. 2025EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Es Attention uses Q matrix to build global model and strengthens spatial differences.•C-Es Attention reduces redundant information in the channel dimension.•We propose a novel segmentation network—Lite-MixedNet.