OBJECTIVE: To construct machine learning prediction model for sepsis-associated encephalopathy (SAE), and analyze the application value of the model on early identification of SAE risk in elderly septic patients. METHODS: Patients aged over 60 years with a primary diagnosis of sepsis admitted to intensive care unit (ICU) from 2008 to 2023 were selected from Medical Information Mart for Intensive Care-IV 2.2 (MIMIC-IV 2.2). Demographic variables, disease severity scores, comorbidities, interventions, laboratory indicators, and hospitalization details were collected. Key factors associated with SAE were identified using univariate Logistic regression analysis. The data were randomly divided into training and validation sets in a 7 : 3 ratio. Multivariable Logistic regression analysis was conducted in the training set and visualized using a nomogram model for prediction of SAE. The discrimination of the model was evaluated in the validation set using the receiver operator characteristic curve (ROC curve), and its calibration was assessed using calibration curve. Furthermore, multiple machine learning algorithms, including multi-layer perceptron (MLP), support vector machine (SVM), naive bayes (NB), gradient boosting machine (GBM), random forest (RF), and extreme gradient boosting (XGB), were constructed in the training set. Their predictive performance was subsequently evaluated on the validation set. Taking the XGB model as an example, the interpretability of the model through the SHapley Additive exPlanations (SHAP) algorithm was enhanced to identify the key predictive factors and their contributions. RESULTS: A total of 2 204 septic patients were finally enrolled, of whom 840 developed SAE (38.1%). A total of 21 variables associated with SAE were screened through univariate Logistic regression analysis. Multivariable Logistic regression analysis showed that endotracheal intubation [odds ratio (OR) = 0.40, 95% confidence interval (95%CI) was 0.19-0.88, P < 0.001], oxygen therapy (OR = 0.76, 95%CI was 0.53-0.95, P = 0.023), tracheotomy (OR = 0.20, 95%CI was 0.07-0.53, P < 0.001), continuous renal replacement therapy (CRRT; OR = 0.32, 95%CI was 0.15-0.70, P < 0.001), cerebrovascular disease (OR = 0.31, 95%CI was 0.16-0.60, P < 0.001), rheumatic disease (OR = 0.44, 95%CI was 0.19-0.99, P < 0.001), male (OR = 0.68, 95%CI was 0.54-0.86, P = 0.001), and maximum anion gap (AG; OR = 0.95, 95%CI was 0.93-0.97, P < 0.001) were associated with an decreased probability of SAE, and age (OR = 1.05, 95%CI was 1.03-1.06, P < 0.001), acute physiology score III (APSIII; OR = 1.02, 95%CI was 1.01-1.02, P < 0.001), Oxford acute severity of illness score (OASIS; OR = 1.04, 95%CI was 1.03-1.06, P < 0.001), and length of hospital stay (OR = 1.01, 95%CI was 1.01-1.02, P < 0.001) were associated with an increased probability of SAE. A nomogram model was constructed based on these variables. In the validation set, ROC curve analysis showed that the model achieved an area under the ROC curve (AUC) of 0.723, and the calibration curve showed good consistency between the predicted probability of the model and the observed probability. Among the machine learning algorithms, including MLP, SVM, NB, GBM, RF, and XGB, the SVM model and RF model demonstrated relatively good predictive performance, with AUC of 0.748 and 0.739, respectively, and the sensitivity was both exceeding 85%. The predictive performance of the XGB model was explained through SHAP analysis, and the results indicated that APSIII score (SHAP value was 0.871), age (SHAP value was 0.521), and OASIS score (SHAP value was 0.443) were important factors affecting the predictive performance of the model. CONCLUSIONS The machine learning-based SAE prediction model exhibits good predictive capability and holds significant application value for the early identification of SAE risk in elderly septic patients.
Humans ; Machine Learning ; Aged ; Sepsis-Associated Encephalopathy ; Sepsis/complications* ; Intensive Care Units ; Logistic Models ; Middle Aged ; Male ; ROC Curve ; Female ; Bayes Theorem ; Nomograms ; Support Vector Machine ; Algorithms

Objective To study the correlation between serum thioredoxin 1(TRX-1)and nucleotide-bind-ing oligomerization domain-like receptor protein 3(NLRP3)expression levels and functional dyspepsia(FD)in children.Methods A total of 45 children with FD who were admitted to Shijiazhuang Maternal and Child Health Hospital from May 2022 to June 2023 were enrolled as the FD group.Another 40 healthy children were selected as the control group.Real-time fluorescence quantitative PCR was used to measure the expression lev-els of serum TRX-1 and NLRP3 mRNA.Spearman correlation analysis was used to analyze the correlation be-tween the expression levels of serum TRX-1 mRNA,NLRP3 mRNA and FD related scores.Multivariate Lo-gistic regression was used to analyze the influencing factors of FD.Results The expression levels of TRX-1 mRNA and NLRP3 mRNA in FD group were higher than those in control group(P＜0.05).The mild and moderate FD groups had significantly lower serum TRX-1 mRNA and NLRP3 mRNA expression levels than the severe FD group,and the mild FD group had significantly lower serum TRX-1 mRNA and NLRP3 mRNA expression levels than the moderate FD group(P＜0.05).The expression levels of TRX-1 mRNA and NLRP3 mRNA in serum were positively correlated with egigastric pain,abdominal distension or epigastric discomfort,nausea and vomiting(P＜0.05),but were not correlated with early satiety,loss of appetite or food intake,and belching(P＞0.05).The expression level of serum NLRP3 mRNA was positively correlated with the total score(P＜0.05),while the expression level of serum TRX-1 mRNA was not correlated with the total score(P＞0.05).The increased levels of serum TRX-1 mRNA and NLRP3 mRNA were independent risk factors for FD(P＜0.05).Conclusion The expression levels of TRX-1 and NLRP3 may play an important role in the pathogenesis of FD.

Objective To investigate the changes and clinical significance of serum monocyte chemoattractant protein-1 (MCP-1) and prostaglandin E₂ (PGE₂) levels in children with functional dyspepsia (FD). Methods A retrospective study was conducted to enroll 89 FD children admitted from October 2022 to October 2023 as FD group, and 91 healthy children in the same period were selected as control group. The FD group was further divided into mild group (n=36), moderate group (n=33), and severe group (n=20) based on the severity of the disease. The serum levels of MCP-1 and PGE₂ were detected by enzyme-linked immunosorbent assay (ELISA). The severity of the disease was assessed by the Global Overall Symptom Score (GOSS), and gastric emptying was evaluated in all the children. Pearson and Spearman correlation analyses were performed to explore the correlations of serum MCP-1 and PGE₂ levels with gastric half-emptying time and FD symptom scores. Results The serum MCP-1 level in the FD group was significantly higher than that in the control group, while the PGE₂level was significantly lower (P < 0.05). In the severe group, the serum MCP-1 level was significantly higher while the PGE₂ level was significantly lower than that in the moderate and mild groups, and similar differences were significantly observed between the moderate and mild groups (P < 0.05). The gastric half-emptying time and postprandial 2-hour residual rate were significantly higher in the FD group than the control group (P < 0.05). Serum MCP-1 in FD children was positively correlated with gastric half-emptying time and postprandial 2-hour residual rate (P < 0.05), while serum PGE₂ was negatively correlated with these parameters (P < 0.05). Serum MCP-1 in FD children was significantly positively correlated with upper abdominal burning sensation, early satiety, belching, upper abdominal pain, and the total GOSS score (r=0.611, 0.509, 0.504, 0.491, 0.513; P < 0.05), while PGE2 was significantly negatively correlated with upper abdominal burning sensation, early satiety, belching, upper abdominal pain, acid reflux, and GOSS total score (r=-0.516, -0.561, -0.493, -0.533, -0.497, -0.482; P < 0.05). Conclusion Elevated serum MCP-1 and decreased PGE₂ levels in children with FD are closely related to the severity of the disease, gastric emptying, and some FD symptoms.

Objective To analyze the prevalence and risk factors of anemia in a general population in Kailuan.Methods Working and retired employees m Kailuan Company who had participated in biennial physical examination from 2006-2014 were investigated by questionnaire and blood test.Hemoglobin levels＜ 120 g/L in male and＜ 110 g/L in female are defined as anemia.The trend of prevalence of anemia was analyzed by chi square test.Multivariable logistic regression was used to analyze the factors associated with anemia.Results (1) The biennial prevalence of anemia in Kailuan during 2006-2014 were 3.7％,3.1％,2.4％,1.3％,1.5％.The corresponding proportion were 3.3％,2.3％,1.9％,0.8％,1.0％ in males and 5.3％,5.9％,4.2％,3.1％ and 3.3％ in females,respectively.The differences between males and females were statistically significant (all P＜0.05).The prevalence of anemia declined over time (P for trend＜0.05).(2) The results of multivariable logistic regression showed that aging and elevated hs-CRP were positively associated with anemia,with OR=1.01 (95％CI 1.01-1.02)and 1.03 (95％CI 1.02-1.03),respectively.While male,BMI,physical exercise,smoking,hyperlipidemia were negatively associated with anemia with OR=0.60(95％CI 0.55-0.65),0.99 (95％CI 0.98-0.99),0.91 (95％CI 0.82-0.98),0.87 (95％CI 0.81-0.95)and 0.87(95％CI0.81-0.94),respectively.Conclusions The prevalence of anemia in a large general population in Kailuan has been analyzed.Prevalence of anemia is higher in males than females and declines over time.Several demographic and clinical characteristics are associated with anemia.