A hybrid attention multi-scale fusion network for real-time semantic segmentation.

In semantic segmentation research, spatial information and receptive fields are essential. However, currently, most algorithms focus on acquiring semantic information and lose a significant amount of spatial information, leading to a significant decrease in accuracy despite improving real-time inference speed. This paper proposes a new method to address this issue.

Effect of GOLPH3 on cumulus granulosa cell apoptosis and ICSI pregnancy outcomes.

Previous studies have shown that GOLPH3 mediates cell growth, proliferation and differentiation and inhibits cell apoptosis; however, the role of GOLPH3 in cumulus granulosa cells and the value of GOLPH3 in predicting ICSI pregnancy outcomes remain unknown until now. Our findings showed higher positive expression rate, score of staining intensity, and immunohistochemical score of GOLPH3 in the cumulus granulosa cells of the pregnant women relative to non-pregnant women, and a higher apoptotic rate of cumulus granulosa cells was detected in non-pregnant women than in pregnant women. Pearson correlation analyses revealed that pregnancy correlated negatively with GOLPH3 expression and apoptosis of cumulus granulosa cells, and positively with the number of follicles punctured, number of grade III oocytes, number of eggs retrieved for ICSI, number of zygotes, number of cleavage-stage embryos, number of top-quality embryos, number of blastocysts, number of top-quality blastocysts, and number of frozen embryos.