Background: Patients with COVID-19 in the intensive care unit (ICU) have a high mortality rate, and methods to assess patients’ prognosis early and administer precise treatment are of great significance. Objective: The aim of this study was to use machine learning to construct a model for the analysis of risk factors and prediction of mortality among ICU patients with COVID-19. Methods: In this study, 123 patients with COVID-19 in the ICU of Vulcan Hill Hospital were retrospectively selected from the database, and the data were randomly divided into a training data set (n=98) and test data set (n=25) with a 4:1 ratio. Significance tests, correlation analysis, and factor analysis were used to screen 100 potential risk factors individually. Conventional logistic regression methods and four machine learning algorithms were used to construct the risk prediction model for the prognosis of patients with COVID-19 in the ICU. The performance of these machine learning models was measured by the area under the receiver operating characteristic curve (AUC). Interpretation and evaluation of the risk prediction model were performed using calibration curves, SHapley Additive exPlanations (SHAP), Local Interpretable Model-Agnostic Explanations (LIME), etc, to ensure its stability and reliability. The outcome was based on the ICU deaths recorded from the database. Results: Layer-by-layer screening of 100 potential risk factors finally revealed 8 important risk factors that were included in the risk prediction model: lymphocyte percentage, prothrombin time, lactate dehydrogenase, total bilirubin, eosinophil percentage, creatinine, neutrophil percentage, and albumin level. Finally, an eXtreme Gradient Boosting (XGBoost) model established with the 8 important risk factors showed the best recognition ability in the training set of 5-fold cross validation (AUC=0.86) and the verification queue (AUC=0.92). The calibration curve showed that the risk predicted by the model was in good agreement with the actual risk. In addition, using the SHAP and LIME algorithms, feature interpretation and sample prediction interpretation algorithms of the XGBoost black box model were implemented. Additionally, the model was translated into a web-based risk calculator that is freely available for public usage. Conclusions: The 8-factor XGBoost model predicts risk of death in ICU patients with COVID-19 well; it initially demonstrates stability and can be used effectively to predict COVID-19 prognosis in ICU patients.

Granulomas are a key pathological feature of tuberculosis (TB), characterized by cell heterogeneity, spatial composition, and cellular interactions, which play crucial roles in granuloma progression and host prognosis. This study aims to analyze the transcriptome profiles of cell populations based on their spatial location and to understand the core transcriptome characteristics of granuloma formation and development.

Background: The rapid outbreak of coronavirus disease 2019 (COVID-19) is a major health concern, in response to which widespread risk factor research is being carried out. Objective: To discover how physical activity and lifestyle affect the epidemic as well as the disease severity and prognosis of COVID-19 patients. Methods: This multicenter, retrospective cohort study included 203 adults infected with COVID-19 and 228 uninfected adults in three Chinese provinces, with 164 (80.7%) of the infected participants and 188 (82.5%) of the uninfected participants answering a doctor-administered telephone questionnaire on lifestyle. The binary logistic regression model and the ordinal logit model were used to observe relevance. Results: Comparing sick and non-sick patients, we found that irregular exercise (P=0.004), sedentary lifestyle (P=0.010), and overexertion (P< 0.001) may be associated with the susceptibility to COVID-19. In symptomatic patients, using the recommended status as a reference, risk of severe infection increased with decreased sleep status, being 6.729 (95% CI=2.138– 21.181) times higher for potentially appropriate sleep (P=0.001) and peaking at 8.612 (95% CI=1.913– 38.760) times higher for lack of sleep (P=0.005). Reduction in average daily sleep time significantly increased the likely severity (P=0.002). Discussion: Through further examination of damage of external lung organs, we found that lack of sleep affected not only disease severity but also prognosis. Based on these findings, the public should prioritize a healthy lifestyle and get adequate sleep in response to the outbreak. The study of life habits may bring new ideas for the prevention and treatment of COVID-19.

Dear Editor,Avian influenza virus (AIVs) is an influenza A virus belonging to the family Orthomyxoviridae which consists of eight single-stranded negative-sense RNA fragments.AIVs can infect a broa...

Eosinophilic asthma (EA) is a common subtype of asthma and often progresses to severe disease. In order to understand its pathogenesis, targeted next-generation gene sequencing was performed on 77 Chinese EA patients and 431 Chinese healthy controls to obtain differential genomic variations. Among the 41 Single Nucleotide Polymorphisms (SNPs) screened for mutation sites in more than 3 patients, filaggrin gene FLG rs192116923 T>G and FLG rs75235053 C>G were newly found to be associated with EA patients with atopic dermatitis (AD) (P G may increase the binding of Smad3 to transmit TGF-β1 signaling, and thereby inhibit filaggrin expression, and FLG rs75235053 C>G may add new splicing sites to reduce filaggrin monomers. It has been known that the level of Th2 cytokine IL-4 is increased in EA patients, and IL-4 increases airway epithelial permeability and enhances inflammatory response through some unclear mechanisms. To figure out whether filaggrin is involved in immune responses in asthma, we have treated human respiratory epithelial cell line BEAS-2B cells with IL-4 and found that the expression levels of filaggrin and E-cadherin decreased significantly in a time and dose-dependent manner, suggesting that IL-4 increased airway epithelial permeability by reducing filaggrin and adhesion molecule. In addition, in our study, IL-4 increased the expression of epithel-derived inflammatory cytokines IL-33 and TSLP which further enhanced the Th2 inflammatory response. To investigate the role of filaggrin in development of EA, knockdown filaggrin with siRNA revealed a decrease in E-cadherin levels, which were further down-regulated by IL-4 stimulation. Knockdown of filaggrin alone did not affect the levels of IL-33 and TSLP, but further exacerbated the decrease of IL-33/TSLP caused by IL-4, suggesting that filaggrin may involve in IL-4R signaling pathway to regulate the level of IL-33/TSLP. In conclusion, in the Th2 cytokine milieu of asthma, FLG deficient mutation in airway epithelial cells may increase the epithelial permeability and the expression of IL-33/TSLP which positively feedback the Th2 inflammation response.