Search

Background: Research on biomarkers associated with the severity and adverse prognosis of COVID-19 can be beneficial for improving patient outcomes. However, there is limited research on the role of soluble TREM-1 (sTREM-1) in predicting the severity and prognosis of COVID-19 patients. Methods: A total of 115 COVID-19 patients admitted to the emergency department of Beijing Youan Hospital from February to May 2023 were included in the study. Demographic information, laboratory measurements, and blood samples for sTREM-1 levels were collected upon admission. Results: Our study found that sTREM-1 levels in the plasma of COVID-19 patients increased with the severity of the disease (moderate vs mild, p=0.0013; severe vs moderate, p=0.0195). sTREM-1 had good predictive value for disease severity and 28-day mortality (area under the ROC curve was 0.762 and 0.805, respectively). sTREM-1 also exhibited significant correlations with age, body temperature, respiratory rate, PaO2/FiO2, PCT, CRP, and CAR. Ultimately, through multivariate logistic regression analysis, we determined that sTREM-1 (OR 1.008, 95% CI: 1.002– 1.013, p=0.005), HGB (OR 0.966, 95% CI: 0.935– 0.998, p=0.036), D-dimer (OR 1.001, 95% CI: 1.000– 1.001, p=0.009), and CAR (OR 1.761, 95% CI: 1.154– 2.688, p=0.009) were independent predictors of 28-day mortality in COVID-19 patients. The combination of these four markers yielded a strong predictive value for 28-day mortality in COVID-19 cases with an AUC of 0.919 (95% CI: 0.857 − 0.981). Conclusion: sTREM-1 demonstrated good predictive value for disease severity and 28-day mortality, serving as an independent prognostic factor for adverse patient outcomes. In the future, we anticipate conducting large-scale multicenter studies to validate our research findings. [ABSTRACT FROM AUTHOR]

Field-programmable gate array (FPGA) has become an excellent hardware accelerator solution for convolutional neural networks (CNNs). Meanwhile, optimizing methods, such as model compression, have been proposed. As most CNN accelerators focus on dense neural networks, to solve the problem of difficult hardware deployment due to irregular networks, we propose a method for sparse neural networks in our work. The storage and coding format of sparse data obtained by the block pruning method is designed to make it friendly to implement on FPGA. Besides, we also propose an efficient and simple data flow by the planarization of the whole convolution calculation process. The experimental result demonstrates that our implementation can achieve clock frequency of 190 MHz, power consumption of 13.32 W and inferencing speed of 16.37 ms. Compared with some typical Mobilenet implementation schemes, our method has been proven to achieve a better balance between frequency, accuracy, power consumption, and speed. [ABSTRACT FROM AUTHOR]

Seasonal temperature reconstructions from ice cores are missing over glacial–interglacial timescales, preventing a good understanding of the driving factors of Antarctic past climate changes. Here the total air content (TAC) record from the Antarctic EPICA Dome C (EDC) ice core is analyzed over the last 440 ka (thousand years). While the water isotopic record, a tracer for annual mean surface temperature, exhibits a dominant ∼100 kyr cyclicity, the TAC record is associated with a dominant ∼40 kyr cyclicity. Our results show that the TAC record is anti-correlated with the mean insolation over the local astronomical summer half-year. They also show for the first time that it is highly anti-correlated with local summer temperature simulated with an Earth system model of intermediate complexity. We propose that (1) the local summer insolation controls the local summer temperature; (2) the latter, through the development of temperature gradients at the near-surface of the ice sheet (<2 m), is affecting the surface snow structure; and (3) those snow structure changes propagating down to the bottom of the firn through densification are eventually controlling the pore volume at the bubble close-off and consequently the TAC. Hence, our results suggest that the EDC TAC record could be used as a proxy for local summer temperature changes. Also, our new simulations show that the mean insolation over the local astronomical summer half-year is the primary driver of Antarctic summer surface temperature variations, while changes in atmospheric greenhouse gas (GHG) concentrations and Northern Hemisphere (NH) ice sheet configurations play a more important role in Antarctic annual surface temperature changes. [ABSTRACT FROM AUTHOR]