SuperMini-seg: An ultra lightweight network for COVID-19 lung infection segmentation from CT images

The automatic segmentation of lung lesions from COVID-19 computed tomography (CT) images is helpful in establishing a quantitative model to diagnose and treat COVID-19. To this end, this study proposes a lightweight segmentation network called the SuperMini-Seg. We propose a new module called the transformer parallel convolution module (TPCB), which introduces both transformer and convolution operations in one module. SuperMini-seg adopts the structure of a double-branch parallel to downsample the image and designs a gated attention mechanism in the middle of the two parallel branches. At the same time, the attentive hierarchical spatial pyramid (AHSP) module and criss-cross attention module are adopted, and more than 100K parameters are present in the model. At the same time, the model is scalable, and the parameter quantity of SuperMini-seg-V2 reaches more than 70K. Compared with other advanced methods, the segmentation accuracy was almost reached the state-of-art method. The calculation efficiency was high, which is convenient for practical deployment.

Automated summary

The study proposes a lightweight segmentation network called SuperMini-Seg for automatically segmenting lung lesions from COVID-19 CT images, which is helpful in establishing a quantitative model for diagnosing and treating COVID-19. The network introduces a new module called the transformer parallel convolution module (TPCB) that combines transformer and convolution operations. It adopts a double-branch parallel structure to downsample the image and designs a gated attention mechanism in between. The model also incorporates the attentive hierarchical spatial pyramid (AHSP) module and criss-cross attention module, with over 100K parameters. Compared to other advanced methods, it achieves almost state-of-the-art segmentation accuracy and high calculation efficiency, making it suitable for practical deployment.