SFIT-Net: Spatial Reconstruction Feature Interaction Transformer Retinal Vessel Segmentation Algorithm

Liming Liang, Baohe Lu, Jian Wu, Yulin Li, Xiaoqi Sheng

Abstract: Highlights•A SFIT-Net method combines CNN and transformer to enhance vessel segmentation.•A FIT module captures long-range dependencies, and SRU reduces vascular info loss.•TFA merges same-scale features at decoding to enhance microvessel capture ability.•The proposed model outperforms state-of-the-art methods with fewer parameters.