MAXFormer: Enhanced transformer for medical image segmentation with multi-attention and multi-scale features fusion

Zhiwei Liang, Kui Zhao, Gang Liang, Siyu Li, Yifei Wu, Yiping Zhou

Abstract: Highlights•Novel transformer module: efficient parallel local–global attention architecture.•Refined Fused Connection module for compensating spatial detail loss.•Enhancing accuracy of medical image segmentation for clinical applications.