Research Reports from Hangzhou Medical College Provide New Insights into Bioengineering (DA-TransUNet: integrating spatial and channel dual attention with transformer U-net for medical image segmentation).

Předmět:	IMAGE segmentation DIAGNOSTIC imaging TRANSFORMER models BIOENGINEERING MEDICAL schools
Zdroj:	Medical Imaging Week; 6/4/2024, p458-458, 1p
Abstrakt:	A recent study conducted by researchers at Hangzhou Medical College in China has proposed a new framework called DA-TransUNet for accurate medical image segmentation. The framework integrates the Transformer and dual attention block (DA-Block) into the traditional U-shaped architecture, addressing issues related to parameter efficiency and computational complexity. DA-TransUNet improves the model's capability to extract features and consistently outperforms existing techniques in medical image segmentation tasks. The research contributes to the development of high-precision automated medical image diagnosis. The codes and parameters of the model are publicly available for further exploration. [Extracted from the article]
Databáze:	Complementary Index
Externí odkaz:	Zobrazit plný text záznamu