IntroductionWith intense deficiency of medical resources during COVID-19 pandemic, risk stratification is of strategic importance. Blood glucose level is an important risk factor for the prognosis of infection and critically ill patients. We aimed to investigate the prognostic value of blood glucose level in patients with COVID-19.Research design and methodsWe collected clinical and survival information of 2041 consecutive hospitalized patients with COVID-19 from two medical centers in Wuhan. Patients without available blood glucose level were excluded. We performed multivariable Cox regression to calculate HRs of blood glucose-associated indexes for the risk of progression to critical cases/mortality among non-critical cases, as well as in-hospital mortality in critical cases. Sensitivity analysis were conducted in patient without diabetes.ResultsElevation of admission blood glucose level was an independent risk factor for progression to critical cases/death among non-critical cases (HR=1.30, 95% CI 1.03 to 1.63, p=0.026). Elevation of initial blood glucose level of critical diagnosis was an independent risk factor for in-hospital mortality in critical cases (HR=1.84, 95% CI 1.14 to 2.98, p=0.013). Higher median glucose level during hospital stay or after critical diagnosis (≥6.1 mmol/L) was independently associated with increased risks of progression to critical cases/death among non-critical cases, as well as in-hospital mortality in critical cases. Above results were consistent in the sensitivity analysis in patients without diabetes.ConclusionsElevation of blood glucose level predicted worse outcomes in hospitalized patients with COVID-19. Our findings may provide a simple and practical way to risk stratify COVID-19 inpatients for hierarchical management, particularly where medical resources are in severe shortage during the pandemic.
Although deep neural networks are highly effective, their high computational and memory costs severely challenge their applications on portable devices. As a consequence, low-bit quantization, which converts a full-precision neural network into a low-bitwidth integer version, has been an active and promising research topic. Existing methods formulate the low-bit quantization of networks as an approximation or optimization problem. Approximation-based methods confront the gradient mismatch problem, while optimization-based methods are only suitable for quantizing weights and could introduce high computational cost in the training stage. In this paper, we propose a novel perspective of interpreting and implementing neural network quantization by formulating low-bit quantization as a differentiable non-linear function (termed quantization function). The proposed quantization function can be learned in a lossless and end-to-end manner and works for any weights and activations of neural networks in a simple and uniform way. Extensive experiments on image classification and object detection tasks show that our quantization networks outperform the state-of-the-art methods. We believe that the proposed method will shed new insights on the interpretation of neural network quantization. Our code is available at https://github.com/aliyun/ alibabacloud-quantization-networks.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.