GCSA-SegFormer: Transformer-Based Segmentation for Liver Tumor Pathological Images DOI Creative Commons

Jiangkun Wen,

Sihua Yang,

Weiqi Li

и другие.

Bioengineering, Год журнала: 2025, Номер 12(6), С. 611 - 611

Опубликована: Июнь 4, 2025

Pathological images are crucial for tumor diagnosis; however, due to their extremely high resolution, pathologists often spend considerable time and effort analyzing them. Moreover, diagnostic outcomes can be significantly influenced by subjective judgment. With the rapid advancement of artificial intelligence technologies, deep learning models offer new possibilities pathological image diagnostics, enabling diagnose more quickly, accurately, reliably, thereby improving work efficiency. This paper proposes a novel Global Channel Spatial Attention (GCSA) module aimed at enhancing representational capability input feature maps. The combines channel attention, shuffling, spatial attention capture global dependencies within By integrating GCSA into SegFormer architecture, network, named GCSA-SegFormer, accurately information detailed features in complex scenarios. proposed network was evaluated on liver dataset publicly available ICIAR 2018 BACH dataset. On dataset, GCSA-SegFormer achieved 1.12% increase MIoU 1.15% MPA compared baseline models. it improved 1.26% 0.39% Additionally, performance metrics this were with seven different types semantic segmentation, showing good results all comparisons.

Язык: Английский

GCSA-SegFormer: Transformer-Based Segmentation for Liver Tumor Pathological Images DOI Creative Commons

Jiangkun Wen,

Sihua Yang,

Weiqi Li

и другие.

Bioengineering, Год журнала: 2025, Номер 12(6), С. 611 - 611

Опубликована: Июнь 4, 2025

Pathological images are crucial for tumor diagnosis; however, due to their extremely high resolution, pathologists often spend considerable time and effort analyzing them. Moreover, diagnostic outcomes can be significantly influenced by subjective judgment. With the rapid advancement of artificial intelligence technologies, deep learning models offer new possibilities pathological image diagnostics, enabling diagnose more quickly, accurately, reliably, thereby improving work efficiency. This paper proposes a novel Global Channel Spatial Attention (GCSA) module aimed at enhancing representational capability input feature maps. The combines channel attention, shuffling, spatial attention capture global dependencies within By integrating GCSA into SegFormer architecture, network, named GCSA-SegFormer, accurately information detailed features in complex scenarios. proposed network was evaluated on liver dataset publicly available ICIAR 2018 BACH dataset. On dataset, GCSA-SegFormer achieved 1.12% increase MIoU 1.15% MPA compared baseline models. it improved 1.26% 0.39% Additionally, performance metrics this were with seven different types semantic segmentation, showing good results all comparisons.

Язык: Английский

Процитировано

0