Enhanced Cross-stage-attention U-Net for esophageal target volume segmentation.

Purpose: The segmentation of target volume and organs at risk (OAR) was a significant part of radiotherapy. Specifically, determining the location and scale of the esophagus in simulated computed tomography images was difficult and time-consuming primarily due to its complex structure and low contrast with the surrounding tissues. In this study, an Enhanced Cross-stage-attention U-Net was proposed to solve the segmentation problem for the esophageal gross tumor volume (GTV) and clinical tumor volume (CTV) in CT images.

Methods: First, a module based on principal component analysis theory was constructed to pre-extract the features of the input image. Then, a cross-stage based feature fusion model was designed to replace the skip concatenation of original UNet, which was composed of Wide Range Attention unit, Small-kernel Local Attention unit, and Inverted Bottleneck unit. WRA was employed to capture global attention, whose large convolution kernel was further decomposed to simplify the calculation. SLA was used to complement the local attention to WRA. IBN was structed to fuse the extracted features, where a global frequency response layer was built to redistribute the frequency response of the fused feature maps.

Results: The proposed method was compared with relevant published esophageal segmentation methods. The prediction of the proposed network was MSD = 2.83(1.62, 4.76)mm, HD = 11.79 ± 6.02 mm, DC = 72.45 ± 19.18% in GTV; MSD = 5.26(2.18, 8.82)mm, HD = 16.22 ± 10.01 mm, DC = 71.06 ± 17.72% in CTV.

Conclusion: The reconstruction of the skip concatenation in UNet showed an improvement of performance for esophageal segmentation. The results showed the proposed network had better effect on esophageal GTV and CTV segmentation.