Structural basis of FpGalNase and its combination with FpGalNAcDeAc for efficient A-to-O blood group conversion.

Transfusion safety and blood typing continue to present significant challenges in clinical practice, including risks of incorrect blood transfusions and blood shortages. One promising solution is the enzymatic conversion of all red blood cell (RBC) types into universal O-type RBCs. However, the major obstacle to this strategy is the relatively low catalytic efficiency of the enzymes involved.

DSA-Net: Infrared and Visible Image Fusion via Dual-Stream Asymmetric Network.

Infrared and visible image fusion technologies are used to characterize the same scene using diverse modalities. However, most existing deep learning-based fusion methods are designed as symmetric networks, which ignore the differences between modal images and lead to source image information loss during feature extraction. In this paper, we propose a new fusion framework for the different characteristics of infrared and visible images.